*ì'.* PLANT-FUNGAL INTERACTIONS DURdNÇsEp …...the putative myc phenotypes identified two lines...
Transcript of *ì'.* PLANT-FUNGAL INTERACTIONS DURdNÇsEp …...the putative myc phenotypes identified two lines...
Ë,Fh¡'FUS.,"þ
*ì'.*
{@*,+^
PLANT-FUNGAL INTERACTIONS DURdNÇsEp l$srVES ICULAR-ARB US CULAR MYC ORqJIZA
DEVELOPMENT: A MOLECULAR APPRøçHI* Ñ
Phillip James Murphy
B.Ag. Sc. (Honours ), Adelaide
Thesis submitted for the degree ofDOCTOR OF. PHILOSOPHY
lnDepartment of Plant Science
Facurtv or AsricitHil.ii,i X'ïä:,l,iJ"rce
sciences
Et.afil
.\,g$i"
October, 1995
(i {J 4 tj;:
DECLARATION
The work presented in this thesis contains no material which has been accepted for the
award of any other degree or diploma in any university or other tertiary institution and,
to the best of my knowledge and belief, contains no material previously published or
written by another person, except where due reference is made in the text.
I give consent to this copy of my thesis, when deposited in the University Library,
being available for photocopying and loan.
P.J. Murphy
August 1995
Acknowledgement
Special thanks must go to my supervisors, Dr. Sally Smith and Dr. Peter Langridge,
not only for their academic direction and help, but also their enthusiasm,
encouragement and humour. It has been a joy to learn from them. Thank you to all my
colleagues at the V/aite Institute and to Jill, Sam, Brett and Suzie whose friendships I
treasure dearly.
Finally, I must acknowledge the very generous help of Angelo Karakousis who
supplied many of the hybridisation membranes used for the mapping experiments and
also assisted in the preparation of the linkage maps.
Table of Contents
1.1
1.2
1.3
1.5
1.6
r.7
1.8
1.9
2.1 Plant and Fungal Genetic Stocks
2.I.1 Plants
2.1.2 Fungi
2.2 Plant Growth Conditions
2.2.1 PottingMedium
2.2.2 Seed Sterilisation
2.3 Plant Growth
2.4 Harvesting of Plant Material
2.5 Assessment of Myconhizal Colonisation
2.6 Other Material
2.7 General Recombinant DNA Methods
2.7.1 Escherichia coli Methods
2.7 .1.1 Bacterial Culnres
Chapter 1
Introduction
General Introduction
Vesicula¡-Arbuscular Mycorrhizas
VA Myconhizas: Stn¡cture and Development X-
Specificity and Recognition in VA Myconhizal Associations
Effects of Mycorrhiza Formation on Phosphorous Uptake and ^rWhole-Plant Metabolism and Physiology
Cellular Responses to Myconhizal Infection /Incompatibillty in Plant-VAM Fungus Interactions
Aims and Significance of this Project
Chapter 2
Materials and Methods
I
2
3
8
13
t6
l8
2T
23
23
24
24
24
25
25
26
26
2.7.1.2 Preparation of E. coli Competant Cells
2.7.I.3 E. coli Transformations
2.7.2 RNA Methods
2.7 .2.1 Isolation of Plant Total RNA
2.7.2.2 Synthesis of Double-Stranded cDNA and Cloning into
î,gttO 28
2.7.2.3 Packaging and Tansfection of Phage Library
2.7.2.5 RNA Gel Blotting, Prehybridisation and Hybridisation
2.7.2.6 Synthesis of Radiolabelled First-stand cDNA
2.7 .3 DNA Methods
2.7 .3.1 DNA Isolations from Plant and Fungal Tissue
2.7.3.2 Plasmid Isolation fromE coli Transformants
2.7 .3.3 Purification of Large-scale Plasmid DNA Preparations
2.7.3.4 Isolation of Bacteriophage DNA
2.7.3.6 DNA Gel Electrophoresis
2.7.3.7 DNA Gel Blotting, Prehybridisation and Hybridisation
2.7.3,8 Restriction Endonuclease Digests
2.7 .3.9 Isolation of DNA Fragments from Electrophoresis Gels
2.7.3.10 Ligations
2.7 .3.I1 DNA Radio-labelling for Hybridisation Sn¡dies
2.7.3.12 Purification of Radio-labelled DNA
2.7.3.13 DNA Sequencing usngTaq DNA Polymerase
2.7.4 Polymerase Chain Reaction (PCR) Methods
2.7.4.1 PCR Amplification of lgtlO inserts
2.7 .4.2 PCR Amplification of Plant Genomic DNA
26
27
27
28
30
30
30
3l
3l
32
33
33
35
35
35
35
36
36
36
36
3.1
3.2
Chapter 3
Growth Response and Infection Studies
of the Hordeum vulgare - Glomus intrardices
Mycorrhizal SYmbiosis
General Introduction
Experiment 1: Variation between tvto Hordeum vulgare cultivars in
Response to Glomus intrardices Colonisation
General Introduction
Bartey Mutagenesis
Mutagenic Agents
Choice of Mutagenic Agent
39
53
54
55
60
60
61
65
3.2.1 Inúoduction
3.2.2 Experimental Methods
3.2.3 Results and Discussion
3.3 Experiment 2: Rate and Extent of VA Mycorrhiza Development in
Glomus intraradices Colonise d Hordeum vulgare cv.'Galleon' Roots
3.3.1 Intoduction
3.3.2 Experimental Methods
3.3.3 Results and Discussion
3.4 Experiment 3: The Distributionof Glomus intraradic¿s Colonisation
in the Roots of. Hordeumvulgare cv. 'Galleon'
3.4.L Introduction
3.4.2 Experimental Methods
3.4.2 Results and Discussion
3.5 Summary and Conclusions
Chapter 4
Isolation of Putative Mycorrhiza-deficient
(myc-) Barley PhenotYPes
43
43
M
4.r
4.2
4.3
4.4
67
68
7l
73
4.5 Generation of M2 Populations from Gamma-ray Inadiated
Hordeum vulgare cv. 'Galleon' seed.
4.6 Screening of the Gamma-iradiated Hordeum vulgare
M2 Populations for rnyc- Mutant Phenotypes
4.7 Polymerase Chain Reaction (PCR) Analysis of the Putative Barley
zyc- Mutants,4-104 and 25-10
4.8 Discussion and Conclusions
Chapter 5
Molecular Cloning and Isolation of
Barley-mycorrhiza Related Plant cDNA's
5.1 Generallntroduction
5.1.1 The Targeted Approach to Gene Cloning
5.L.2 The Non-targeted Approach to Gene Cloning
5.2 Construction, Characterisation and Differential Screening of a
cDNA library Prepared from Mycorrhizal Hordewn vulgare cv.
'Galleon'Roots
5.2.t Synthesis of cDNA
5.2.2 Constn¡ction and Characterisation of a M+ cDNA Library
5.2.3 Isolation of Putative Barley-mycorrhiza Related (BNR)
cDNA clones
5.2.4 DNA:RNA (Northern) Hybridisation Analysis of Putative BMR
cDNA clones
5.2.5 DNA:DNA (Southern) Hybridisation Analysis of the two cDNA
clones BMR6 andBMRTS
5.3 Discussion
74
76
83
89
91
92
92
ll6
t2r
97
97
99
108
rl4
Chapter 6
Molecular Characterisation of
the Barley-mycorrhiza Related cDNA's:
BMR6 and BMR78
6.1 Generallntroduction
6.2 Subcloning of BMR6 and BMR78 into the Plasmid Vector pTZl8U
6.3 Expression of the BMR6 and BMR78 clones
6.3.1 Aims
6.3.2 Plant Growth
6.3.3 ExpressionAnalysis
6.3.4 Results and Discussion
6.4 Chromosomal Mapping of the oDNA Clones BMR6 and BMR78
6.4.1 Aims
6.4.2 Genetic Mapping
6.4.3 AdditionLine Mapping
6.5 Sequencing of the cDNA clones BMR6 and BMR78
6.5.1 Aims
6.5.2 Sequencing Reactions
6.5.3 Sequence Analysis of BMR6 andBMRTS
6.6 Discussion
Chapter 7
General Discussion
124
125
t26
t27
t28
t28
r33
133
135
140
140
140
145
7.r
7.2
7.3
7.4
Model Plant-fungus Systems for Investigating Mycorrhizal Symbioses
Towards Identifying Myc Mutants
Molecular Approaches to Cloning Functional Mycorrhiza Genes
Concluding Statements
148
149
150
151
Appendix 1
Appendix 2
Appendix 3
Appendix 4
Appendix 5
Appendix 6
References
Appendices
Sources or Composition of Materials and Solutions
Protocol for DNA Isolations from Plant Roots and
Fungal Spores
Protocol for DNA Isolations from Plant I-eaves
Prorocol for Plasmid DNA Isolations
Protocol for Phage DNA Isolations
PCR-primer Sequences
153
186
r92
194
t96
198
200
t4d',4lïh tp.f,i Fti $ ¡"i nnÅAY
Plant-fungal Interactions during Vesicular-Ar Uhi-'ji:fi $ lïl û$r,¡r il [L:\ì iÌ iì
Mycorrhiza Developmenû A Molecular Approach.
Phillip James Murphy
Vesicular-a¡buscular (VA) mycorrhiza formation is a complex process which is
under the genetic control of both plant and fungus. It has proved difficult to elucidate
the mechanisms which mediate the formation and maintenance of these long term
compatible symbioses. ExperimeÍrtal evidence suggests that host plant cells are
cytologically modified in direct response to infection by VA mycorrhizal fungi. It is
reasonable to suggest that this is a result of altered gene expression within host-plant
cells.
The aims of the project were; to develop a model infection system in Hordeum
vulgare L. (barley) suitable for undertaking a molecular analysis of the symbiosis; to
identify host plant genes differentially expressed during the early stages of the
infection process; and to produce and screen a mutant barley population for
phenotypes which form abnormal mycorrhizas.
A model infection system was developed using H. vulgare cv. 'Galleon' and the VA
mycorrhizal fungus Glotnus intraradices Schenk and Smith. High levels of early-
infection stage mycorrhizas (20-35Vo of the total root length) were routinely obtained
in the roots of L2-14 day old barley seedlings. Detailed infection studies showed that
the majority of the infection occurred in the first 5-10 lateral roots of each primary
root. The level of infection within these roots was as high as 75Vo of the total root
length.
Complementary DNA (cDNA) libra¡ies were prepared in the bacteriophage vector
Àgt10 from reverse transcribed poly(A) tailed RNA's extracted from 12 day old
mycorrhizal barley roots. Differential screening of the phage library with radio-
labeled single stranded cDNA's (reverse transcribed from nonmycorrhizal and
mycorrhizal lateral root RNA), identified over 500 differentially expressed clones.
Further hybridisation studies, using total RNA extracted sequentially during VA-
mycorrhiza development, confirmed that several of the barley mycorrhiza-related
(BMR) clones represent RNA species which increase or decrease in abundance
during mycorrhiza formation.
Two clones, BMR6 and BMR78, were cha¡acterised in greater detail. These clones
detect RNA transcripts that accumulate to much higher levels in mycorrhizal roots
than in nonmycorrhizal root tissues. The levels of these transcripts within the roots
did not appear to change in response to infection by virulent or avirulent strains of
the root pathogen Geaumanomyces graminis or variable phosphate nutrition but
may have been slightly increased by droughting the plants. DNA hybridisation
analysis confirmed that both of the clones were of plant rather than fungal origin.
Hybridisation studies and sequence analysis strongly suggest that BMR78 represents
a low copy proton-ATPase gene which is located on the long arm of barley
chromosome 2. B.Ml6 detects a low copy gene located on the short arm of barley
chromosome 6; its function is less clear. The DNA sequence of BMR6 has regions,
which have homologies to 2 stress induced genes isolated from Escherichía coli and
a senescence-associated gene isolated from Raphanus sativus L.. The implications
of these results are discussed with respect to available cytological evidence.
\Well characterised mutant plant phenotypes have proven essential to the
understanding of many physiological, biochemical and developmental processes. To
complement this investigation of plant gene expression during VA mycorrhiza
development, two populations of mutagenised Galleon barley were generated by 60-
Co irradiation treatment. The screening of approximately 1850 M2 plants identified
several lines which appeared to be VA mycorrhizal (myc ) mutants. Rescreening of
the putative myc phenotypes identified two lines (MF25-10 and MF4-10a), that
were less susceptible to mycorrhizal infection than the parental Galleon genotype.
Plants of the MF25-10 line developed normally except for slightly lower levels of
otherwise-normal VA mycorrhizas when compared with the parental Galleon. The
other putative myc mutant, MF4-10a, displayed unusual shoot and root development
and formed very few VA mycorrhizas. All plants from line MF4-10a died from what
appeared to be nutrient deficiency within 25 days of germination; no genetic
analyses were carried out, but the segtegation data available indicates that this lethal-
myc phenotype is inherited as a recessive trait. It is proposed that an induced
mutation affecting the visco-elastic properties of line MF4-10a cell walls (as
evidenced by thickened root morphology) could be responsible for rendering the root
cells not only impermeable to soil solutes but almost inpenetrable to G. intraradices.
I
Chaoter l
Introduction
1.1 General Introduction
The term mycorrhiza describes tissue that a¡ises from several distinct symbiotic
associations which occur between fungi of the Glomales (Zygomycetes) and the roots
of most higher plants. These associations are mutualistic (Lewis, 1973) rather than
parasitic; no disease symptoms are obsen¡able, and indeed, the growth of the plant is
very often enhanced by the interaction. In essence, these root stn¡ctures represent the
formation of an extensive and intimate interface between a fungus and a higher plant
which is then maintained for the life of the plant.
The classification of mycorrhizas is largely based on the general morphology of the
tissues. Traditionally, they have been classified into two groups; ectomycorrhizas and
endomycorrhizas. In ectomycorrhizas the interface arises from the establishment of a
fungal sheath which completely surrounds the root epidermis and an intercellula¡ net of
fungal hyphae growing between the cells of the plant-root cortex. No intracellular
penetration of plant-root cells is known to occur. On the other hand, in the
endomycorrhizas there is intracellular penetration into plant-root cortical cells by the
fungal hyphae. The resultant intracellula¡ interface being formed between modified
hyphae (coils or much-branched arbuscules) and the invaginated cortical cell
plasmalemma. An intercellular network of fungal hyphae may also arise. Ecto- and
endomycorrhizas both form an extensive web of external hyphae.
Five main groups have been proposed and generally adopted to describe the diversity
of mycorrhizal associations (Smith, 1980). Ectomycorrhizas remain as previously
)
classified whilst the endo- (and the sub-group ectendo-) mycorrhizas have been
reclassifred as vesicular-arbuscular, ericoid, orchidaceous and arbutoid mycorrhizas.
The delineation between types of mycorrhizal associations is by no means distinct but
it has been argued that this system more fully emphasises structural and physiological
similarities (and differences) between the groups. For a more complete discussion on
the basis and details of mycorrhizal classification see the reviews by Lewis (1973,
1975), Smith (1980) and Harley and Smith (1983).
1.2 Vesicular-Arbuscular Mycorrhizas
Fungi capable of entering into mycorrhizal associations with higher plants are found in
all the taxonomic gtoups. The genera from the family Glomales (Morton and Benny,
1990) form morphologically distinctive tissues termed vesicular-arbuscular (VA)
mycorrhizas. These appear to be the commonest and most widespread of the
mycorrhizal fungi.As the experimental workpresented in this thesis only involved VA
mycorrhizal fungi, my discussion and comments shall only pertain to these types of
mycorrhizas unless otherwise stated.
The great majority of terrestial plants, including most agricultural and hoficultural
species, can enter into symbiotic associations with VA mycorrhizal fungi. As a result
of infection, the plant may show greater stress tolerence or disease resistance and
enhanced growth, particularly in soils of moderate or low nutrient status. Indeed,
some horticulturally important tree species seem to have an almost obligate requirement
for VA mycorrhizal associations in order to attain satisfactory growth (see Harley and
Smith, 1983).
The beneficial consequences associated with VA mycorrhiza formation have stimulated
considerable resea¡ch to determine if these fungi can be manipulated on an agtonomic
scale. It may be possible to improve plant growth and soil status whilst reducing the
use of expensive and envi¡onmentally damaging chemical inputs.
3
A major stumbling block in the effective management of these beneficial micro-
organisms, at both the experimental and field level, is the inability to culture VA
mycorrhizal fungi axenically. The organisms are obligate biotrophs which suggests
that specific plant gene products, possibly resulting from chemical signalling between
the symbiotic partners, are required to mediate and maintain fungal growth and
differentiation. Identifying the types of molecules involved may facilitate effective
manipulation of these fungi. Investigation of plant gene expression in mycorrhizal
roots is one approach to this particular problem. This introductory chapter aims to
outline the salient features of VA mycorrhizas (predominantly from a host-plant
perspective) with a view to using a molecular genetic approach to isolate and identify
plant genes controlling the development of the symbiosis.
1.3. VA Mycorrhizas: Structure and Development
Investigating plant-gene expression in mycorrhizal roots requires a clear understanding
of the developmental events which give rise to an effective VA mycorrhiza.
Consideration of the infection process also provides the framework for predicting the
types and expression patterns of plant genes which may be of importance during VA
mycorrhizal formation.
The infection process is a well defined series of events. In order to reflect the natural
course of events, it seems logical to discuss mycorrhiza development in terms of three
separate regions; the rhizosphere where there is no direct contact between the fungus
and plant cell walls; the root surface where initial contact is made; and within the plant
root where contact is intimate and appears continuous.
(Ð The rhizosphere
Infection of plant roots by VA mycorrhizal fungi is initiated by hyphae growing from
neighbouring mycorrhizal roots or soil-borne propagules such as spores or root
segments containing live hyphae (Tommerup and Abbott, 1981; Friese and Allen,
1991). Attempts to culture VA mycorrhizal fungi axenically have so far proved
4
fruitless although viable spores of VA mycorrhizal fungi will readily germinate in
moist sterile soil or on water agar plates. Intact plant roots do not appe¿u to be
necessary for initiation of hyphal growth but enhanced germination in the presence of
plant-roots or their exudates has been observed for VA mycorrhizal fungi (Powell,
1976; Gianninazi-Pearson et a1.,1989; Beca¡d et a1.,1992).
One or more germ-tubes may emerge from each spore but hyphal growth is restricted
to only a few centimetres. Germ-tube growth invitro can be altered by the addition of
a variety of nutrient substrates to the medium. Many organic substrates, vitamins, and
sulphur compounds have been found to promote growth while others such as
magnesium and zinc can inhibit growth (Hepper, 1979, 1984a and 1984b). The
observation by Mosse (1959) that spore reserves a¡e unlikely to be limiting during
early hyphal growth suggests that lack of nutritional sources in the rhizosphere (ot in
vitro) are of minor importance to the continued hyphal growth of the free living
mycelium. It appears likely then, that specific growth factors are required to ensure
continued hyphal growth in freshly germinated spores.
The existence of these proposed growth factors is yet to be established but it seems
reasonable to speculate that they are probably of plant origin. This is supported by the
findings that roots (Mosse, 1988; Beca¡d and Piche, 1989 and 1990), root exudates
(Graham, 1982; Elias and Safu, L987), cell suspension cultures (Carr et aI,1985) and
cell products (Paula and Sequeira, 1990) all stimulate hyphal growth. Fungal
morphogenesis in the rhizosphere prior to root-contact is induced by the roots of host
plants but not non-host plants (Giovannetti et a1.,1993; Giovannetti et al.,l994aand
1994b). Biologically active soil extracts, dialysates or filtrates also influence
germination and early hyphal growth (Mosse,1959; Daniels and Graham, 1976; see
Hepper 1984b) so microbial sources cannot be ruled out. More recently, CO2 and
flavanoid-like compounds have been implicated in the infection process @ecard and
Piché, 1989; Becard et aI., 1992); these results are dealt with more thoroughly later on
in this introductory chapter.
5
(ü) The root surface
'When the fungal hypha eventually makes contact with a plant root a swollen
appressorium may be produced (Garriock et aI., 1989 ) from which an infection peg
usually arises. It seems clear that at this point 'recognition' of the host by the fungal
partner occurs (see Smith and Gianinazzi-Pearson, 1988). It has been proposed that
structures at the root surface may promote firmer attachment (appressorium formation)
by the fungus prior to penetration of the root cells (see Harley and Smith, 1983;
Anderson, 1988).
An infecting hypha may enter the root between or through epidermal cells or (more
rarely) the root hair cells. This appears to occur anywhere o-lonS
the root (Brundett et al., 1985; Smith et aI., 7992).
(iü) V/ithin the plant root
Once inside the root, hyphae grow through into the outer cortical cells where they
branch and begin to spread longitudinally along the root. Frequently, two or more
appressoria occur at discrete sites on a single root. These "infection units" (Cox and
Sanders, 1974) spread through the epidermis and outer cortex with a fan-shaped
morphology and may eventually overlap to form what appears to be a continuous
infection. This phase of hyphal growth may be intercellular or intracellular with the
formation of hyphal coils (Kinden and Brown, 1975). Hyphal spread in the outer
cortex is usually restricted and most fungal growth occurs within the central and inner
cortex of the root (Harley and Smith, 1983).
The mechanism of fr¡ngal penetration, mechanical and/or enzymatic, into epidermal or
cortical plant cells is not known. Cox and Sanders (1974) were first to describe
bulging of the cell wall a¡ound penetrating hypha. More recent ultrastuctural analysis
has shown a clear breakdown of fibrillar components in the outer layers of the
epidermal cell wall at the point of penetration whilst the inner layers appear to distend
and envelop the invading hypha (Bonfante-Fasolo et al., 1981; Bonfante-Fasolo and
6
Grippiolo, l9S2). These results imply that penetration may be a result of mechanical
pressure coupled with elasticity or extensibility (possibly induced) of the cell wall
(Harley and Smith, 1983). Covalent cross-linking of the plant cell wall protein,
extensin (Lamport, 1975), has been implicated with cell wall construction and rigidity
(Chen and Varner, 1985). It has been postulated that proteolytic activity may be
responsible for reducing the rigidity or disrupting the synthesis or assembing of this
protein matrix (Harley and Smith, 1983; Anderson, 1988). This could result in an
altered or incomplete cell wall strucn¡re amenable to distension and hyphal penetration.
VA myconhizal infection pegs appear analogous to similiar stn¡ctures that have been
observed in many other fungal species; modified morphology (usually narrowing of
the hypha) appears to be important in the infection process.
Pectinase activity has also been suggested as a mechanism to facilitate hyphal
penetration through the epidermis (Garcia-Romero et al., 1991). Production of cell
wall-degrading enzymes (cellulases and pectinases) has been demonstrated in vitro
with some ericoid and ectomycorrhizal fungi (Lindeberg and Lindeberg,l9TT; Giltrap
and Lewis, t982; Cervone et al, 1988). There have been no reports demonstrating
cellulase or pectinase activity in planta for VA mycorrhizal fungus.
It is within the cells of the inner cortex that the typical VA mycorrhizal structures,
arbuscules and vesicles, are formed. Penetration of these cells starts as a constricted
hyphal peg which causes the cortical cell wall to invaginate inwards. The invading
hypha regains its normal size after penetrating the cell (Holley and Peterson, 1979) and
immediately begins to branch dichotomously to form a distinctive'tree-like'structure.
The deposition of an amorphous layer or collar of host wall material around the point
of penetration has been described (Dexheimer et aI., 1979; Harley and Smith, 1983;
Bonfante-Fasolo, 1988a and 1988b). The cell wall becomes thinner and eventually
discontinuous as the hyphae grow inwards, but the fungus does not penetrate nor
appear to alter the cell plasmalemma (Bonfante-Fasolo et a1.,1981). These hyphal
'branches' eventually occupy much of the cell volume and form a very extensive
7
interfacial zone with the plant. This zone, comprised of fungal outer wall, an interfacial'n ra.*n'<
G^,O"5U¡ of host origiþa the cell plasmalemma, is thought to facilitate the bidirectional
transfer of nutrients between fungus and plant, a feature of the VA mychorrhizal
symbiosis (see extensive reviews by Gianinazzi-Pearson and Gianinazzi,l9S3; Harley
and Smith, 1983; Smith and Gianinazzi-Pearson, 1988). Not surprisingly,
considerable interest has been centred on this zone and in particular thg c-ompgsilql
and origin of the interfacial matrix in the sea¡ch to elucidate the mechanism of nutrient
transfer.
The fungal wall at the arbuscula¡ interface is much simplified. It is a thin amorphous
-¿.¿<b,czllwla(structure whilst the
- - - 'fungal wall is thick and fibrilla¡ (Bonfante-Fasolo and
Grippiolo, 1982). The matrix appears to be a disorganised wallJike structure as it
contains scattered fibrillar components and polygalacturonic acid (Dexheimer et al.,
t97 9; Bonfante-Fasolo ¿r al, 1990).
The active life of the a¡buscule is estimated to be only 4-15 days @evege and Bowen,
1975; Cox and Tinker, 1976; Alexander et al., 1989). As they senesce' the hyphae
become vacuoalated and eventually empty. Finally, the arbusqule collaPses and
aggregates. The ultrastructural organisation of the fungal walls appears to retairfhe
same cytochemical cha¡acteristics as extracellular hyphae, although there is some
suggestion of degradation (Bonfante-Fasolo et al, l98t). Absorption of the
degenerating arbuscule or some of its contents by the plant cell, which remains active
and reinfectable (see Harley and Smith, 1983) has been suggested as another
mechanism of nutrient transfer (Scannerini, et al., 1975). Cox and Tinker (1976)
dismiss this suggestion as the senescing arbuscule is unlikely to provide enough
phosphorous. The collapsed arbuscule may ultimately become encased in a
polysaccharide fibrillar membrane, probably of plant origin (Dexheimer et aI., 1979).
Intercellular hyphae show similar cytological changes, the hyphae become increasingly
vacuolated with age and may become compartmentalised by the formation of septa
(Harley and Smith, 1983).
8
Most VA mycorrhizal fungi form vesicles within or between the cortical cells which
appear as large terminal swellings on inter- or intracellular hyphae. They contain lipid
droplets (Holley and Peterson, 1979) and usually develop very late in mycorrhizal
formation. They are probably storage structures.
The latter stages of mycorrhizal development are also associated with extensive growth
of hyphae external to the root (Mosse and Hepper, 1975). This is presumed to be in
response to the increased availability of nutrients which have been acquired from the
host. These external hyphae have at least two major roles. They are intrinsic in the
uptake and translocation of nutrients which may eventually be transferred to the host
plant and they serve as a source of secondary infection along and between roots. For
some VA mycorrhizal fungi, spores and vesicles can be produced on the external
hyphae (Harley and Smith, 1983).
The infection of plant roots by VA myconhizal fungi is complex but appears well
defined both temporally and spacially. This suggests that it is highly regulated and may
be at least partly under plant genetic control .
1.5 Specificity and Recognition in VA Mycorrhizal Associations
Plants come into contact with vast numbers of micro-organisms during their lifecycles.
The great majority of these encounters have no observable effect on the growth of the
plant and yield or on reproductive ability because most potential invaders are only
capable of infecting one or a few host species. Microbes inoculated onto non-host
species are usually unable to breach the preformed physical ba¡riers of the plant. If
they do, the plant will often exhibit a resistance response which effectively halts the
growth of the organism within the plant. Interactions characterised by the inability of a
particular micro-organism to overcome the plant physical or biochemical barriers and
colonise the host-plant tissue are usually termed incompatible associations.
9
Plant-microbial interactions which do result in extensive growth of the organism
within the plant a¡e termed compatible: they may be mutualistic, in which case both
plant and micro-organism derive some benefit from the association, or pathogenic. The
outcome of contact between a plant and a micro-organism appe¿us to be at least partly
dependent on a mutual perception between the interacting cells. It is presumed that
specific extracellula¡ molecules of each organism interact and that these events trigger
cellular responses. The means by which these signals a¡e transmitted and transduced
within the cell are also poorly understood.
The concept of cell recognition arose because of the need to formulate a mechanistic
basis for the discriminatory processes which appear to determine compatibility and
possibly the nature of a particular symbiosis. Clarke and Knox (1978) have defined
recognition as the "initial event in cell-cell communication that elicits a biochemical,
physiological or morphological response". Sequeira (1984) suggested that it is "an
early specific event that triggers a rapid, overt response in the host, either facilitating or
impeding further growth of the pathogen". Defining in precise molecular terms the
"initial" or "specific" event in any plant cell interaction has not proved possible to date
and the mechanisms of plant cell communication a¡e still poorly understood.
Molecular recognition and signalling resulting in defined cell reponses are probably
fundamental to many plant biological processes; this includes cell division, cell growth
and development, fertilisation and stress responses. Most researchers investigating
these phenomena have approached the problem from a plant pathologists perspective,
usually with the assumption that incompatibility (the resistance reaction) requires
specific recognition events. Most reviews on the subject (see Sequiera,l9S4i Collinge
and Slusaranko, 1987; Lamb et a1.,1989 and Dixon and La¡nb, 1990; Dixon et al.,
1994) reflect this bias. The reasons may be largely practical; pathogenic interactions
are genetically well characterised and resistance responses are largely dominant gene-
for-gene interactions. The more obvious benefits that are associated with plant
10
pathogen-related research (such as minimising agricultural losses) probably weigh
largely as well.
The more complex the interaction, the greater the probability that a number of events
determine the final outcome. It is likely that a succession of events; trophic or tactic
responses; specific.attachment molecules of host and/or microbial origin; and sensing
or activational mechanisms, promote and define a compatible association. It should be
noted that in the mutualistic Rhizobium -legume symbiosis, a series of recognition
steps appear to condition compatibility (or susceptibility) to infection (see Bauer, 1981;
Nap and Bisseling, 1990).
VA mycorrhizas are fundamentally different from other well characterised plant-
microbial interactions (mutualistic and parasitic) in that they are persistent biotrophic
interactions of very low or broad specificity. This implies that either a generalised or
shared mechanism facilitating compatability has evolved in the majority of plant
species or the plant simply does not 'recognise' the invader and a resistance
(incompatability) response is not activated. The molecula¡ basis of the processes
controlling initial recognition and compatibility between plants and VA mycorrhizal
fungi has engendered much discussion (see Harley and Smith, 1983; Bonfante-Fasolo
and Spanu, 1992) but very little experimental data. The only clear evidence of
recognition between VAM fungi and the potential host plant is the formation of the
fungal appressorium. (Smith and Gîaninazzi-Pearson, 1986). Transcriptional
activation resulting in cutinase gene expression and subsequent penetration of the plant
cuticle barrier has been reported in fungal pathogens including Fusarium solani pisi
(Woloshuk and Kolattukudy, 1986). It seems likely that expression of VAM fungal
genes which control appressorium development, possibly by transcriptional activation,
must occur very quickly after contact between the two partners. Clearly it is a
challenge to elucidate these molecular events.
11
Certain phenolic compounds which induce gene expression in plant-associated bacteria
have been identified. Molecular analysis indicates that wound-induced monocyclic
plant phenolics, particularly acetosyringone and some related compounds including
chalcones, induce the expression of several virulence genes of Agrobacterium
tumefaciens (Stachel et aI, 1988; Bolton et aI., 1986). Expression of these genes is
necessary for T-DNA transfer and subsequent tumour formation.
Similarly, one particula¡ class of plant phenolics, the flavonoids, induce Rhizobium
nodulation (nod ) genes which are required for the formation of effective symbiotic
associations with legumes (see Long, 1989a and 1989b). These compounds, some of
which have been shown to be host-symbiont specific (Rossen et aL,1985 and 1987),
are only found in the root exudate from zones where root hairs are emerging
(Redmond et aI, t986). These zones are developmentally responsive to Rhizobium
infection (Bhuvaneswari et aI,l98l;Bhuvaneswari and Solheim, 1985). This type of
evidence (see Innes et a1.,1885; Rolfe, 1988 and Rolfe and Gresshoff, 1988) shows
that in at least some highly evolved symbiotic (parasitic or mutualistic) systems,
signalling between the plant and microbe occurs before intimate or even direct contact
is made.
There has been growing speculation that plant phenolic compounds may play a similar
role and act as regulators of gene expression in VAM fungi-plant associations
(Anderson, 1988; Becard and Piche, 1989; Peters and Verma,l99O; Phillips and Tsai,
1992). Some recent experimental evidence supports this contention.
The commercially available flavonones, hesperitin and naringenin, and the flavone
apigenin were shown to enhance in vitro spore germination and./or hyphal growth of
Gigasporø margaritaby Gianiazzi-Pearson et al. (1989). The physiological relevance
of these results is unclear, as these compounds had no significant effect on VA
mycorrhiza formation in white clover (Trifolium repens L.) grown with two Glomus
sp. (Siqueira et aI,1991). However, Siqueira et aI. (1991) were able to demonstrate
T2
that two isoflavanoids, formononetin and biochanin A, which had been isolated from
white clover roots (Nair et a1.,1991) markedly increased VA mycorrhiza formation.
Almost nothing is known about the modifications in plant gene expression which
occur during the establishment of the symbiosis. Root cell-wall-bound peroxidase and
chitinase activities appear to increase in leek seedlings during the early stages of
mycorrhizal infection (Spanu and Bonfante-Fasolo, 1988; Spanu et a1.,1989), whilst
the concentrations of phenolic compounds which have been correlated with
antimicrobial activity (Mansfield, 1983) do not (Cordignola et aI., 1989). The
isoflavanoids, glyceollin, coumestrol and daidzein rapidly accumulated tn Rhizoctonia
solani infected soybean roots, whilst de novo synthesis of these well characterised
phytoalexins was only weakly induced by the mycorrhizal lungi Glomus mosseae and
Glomus fasciculatus (Morandi et aI., 1984; Wyss et al., 1990).
Gianinazzi-Pea¡son et aI. (1990) compared the expression of the pathogenesis related
protein gene, PR-b I (Van Loon et al., 1985) in G. mosseae - and Chalara elegans
-infected tobacco roots. This set of experiments were somewhat flawed because the
authors used different membranes for the comparative total-RNA:PR-b1 cDNA
hybridisations instead of loading all treatments on one filter. However, these results
did indicate that the accumulation of. PR-b I mRNA transcripts was higher in
mycorrhizal roots than in the non-mychorrizal conúols but very low in comparison to
the pathogen infected roots. Franken and Gnadinger (1994) have detected similar
responses in parsley. The protein was detectable using monoclonal antibodies in the C.
elegans infected roots but the levels were either too low to detect or the mRNA's were
not translated in the mycorrhizal roots. Using in situ hybridisation techniques,
Ha¡rison and Dixon (1994) have provided evidence that the expression of Medicago
trunculata genes involved in the flavanoid/isoflavanoid pathway is elevated in (and
only in) Glomus versiþrme -colonised root cotical cells.
Nearly all research into recognition and compatibility in the VAM fungi-plant
symbiosis has been somewhat derivative, taking ideas from the results obtained from
13
comparatively well studied plant-pathogen systems. It appears that plant phenolic
compounds may activate or stimulate early hyphal growth in the rhizosphere and that
(at least some) VA mycorrhizal fungi are capable of inducing a weak or transient
general plant defence response during the early stages of infection. The generality of
these observations remain undetermined and they may be of little biological
significance; the flavanoid concentrations used in the spore germination/hyphal growth
experiments were much higher than physiological levels; and the weak and apparently
localised pathogenJike responses may simply be causally-related to harvesting stress
or damage and the mechanical nature of fungal penetration into the root epidermis.
1.6 Effects of Mycorrhiza Formation on Phosphorous Uptake
and Whole Plant Metabolism and Physiology
Ion uptake by plant roots has two major determinants; the movement or mobility of
ions through the soil matrix and the absorption rate of the root (Nye and Tinker,
lg77).Ions such as NO3-, SO42- and Ca2+ can move rapidly by mass flow through
the soil to the root surface. Uptake is limited by the plant root absorptive capacity. On
the other hand ions such as H2PO4-,NII4+, kP+ and Cu2+ are relatively immobile
and move to the root surface by diffusion. In soils with low P, depletion zones can
rapidly develop around the root zone (see Bieleski and Ferguson, 1983).
Mycorrhizal roots appear to be much more efficient at sequestering soil P than non-
mycorrhizal roots (Cress et al., 1979) because the external VA mycorrhizal hyphae
increase the effective root surface area (Tinker, I975a) absorb P from beyond the
normal depletion zone and a¡e able to translocate it to the plant root (see Bieleski and
Ferguson, 1983). Six fold increases in the rate of P uptake have been reported
(Sanders and Tinker, 1973).
Translocation studies using labelled N sources only accessible to the fungal hyphae
show that VA mycorrhizal celery plants can also accumulate more l5-N than non-
mycorrhizal controls (Ames et a1.,1983; Ames et al , 1984). Smith et aI , (1986)
l4
provided corroborative evidence that external hyphae may also translocate soil N when
they reported greatly increased N inflow into VA mycorrhizal leek plants. This aspect
of nutrient acquisition in VA mycorrhizal roots remains poorþ resea¡ched.
The development of functional mycorrhizas usually increases the uptake of P into host
plant roots. The absence of mycorrhizas can result in P deficiency in non-mycorrhizal
control plants. Direct comparisons between mycorrhizal and non-mycorrhizal plants
are difficult, if not impossible to interpret because many of the growth responses
observed in plants as a result of VA mycorrhiza formation (see Table l) are directly or
indirectly related to improved nutrition (Mosse, 1973) particularly P (Tinker, 1975b)
but also 7-n2+ and Cu2+ (Gilmore, l97L; Kleinschmidt and Gerdemann;Lartbert et
al., 1979).
There is a paucity of information on the effects of P deficiency on specific plant
metabolic pathways (Smith and Gianinazzi-Pearson, 1988). This notion can be
expanded to include nutrient availability in general and needs to be kept firmly in mind
when designing experiments to elucidate the mechanisms controlling plant responses
to VAM fungal infection, particularly at the cellular level where small and/or temporary
changes in P availability may have dramatic and confounding effects.
There are several excellent reviews (see Hayman, 1982; and Smith and Gianinazzi-
Pearson, 1988) detailing the multitude of effects that VA mycorrhiza development has
on whole-plant growth and development. Examples of these effects are presented
below (Table l) in order to emphasise the very broad range of modifications that can
occur and to highlight the fundamental problem associated with comparative
experimentation.
15
Table I Reported host responses to myconhizal infection
Effect on plant
meøbolism/physiology Host olant Reference
Increased photosynthesis
Increased nitrate reductase
and glutamine syntha^se
activities
Increased hydraulic
conductivities
Decreased susceptibility
to wilting
Altered amino acid content
Increased tolerance to root
pathogens
Increased fresh
weight:dry weight ratio
Viciafaba
Allium porrwn
Boutelotn gracilis
TriþIium sp.
Kucey and Paul (1982).
Snellgrove et al., (1986)
Allen et al., (1981).
Oliver et aI., (1983).
Carling et aI., (1978).
Smith et aI , (1985)
Allen et aI., (1982).
Safu ¿r aI., (1971).
Hardie and Layton (1981).
Levy and Krikun (1980).
Nelson and Safir (1982).
I
ll
Bouteloua gracilis
Glycine max
Trifoliwn pratense
Citrus jambhiri
Alliurn porrum
Increased cytokinin levels Boutaloun gracilis Allen et aI., (1980)
Tzamays
Citrus sp.
Glycine max
several species
Young et aI., (1978)
Nemec and Meredittt (1981)
Pacovsky (1989).
Bartchi et al., (1981)
Schenck et aI., (1974 and 1975)
Snellgrove et aI., (1982).
Tester et al , (1987).
Lower root:shoot ratios many species Harley and Smith (1983).
AIIium porum
16
1.7 Cellular Responses to Mycorrhizal Infection
Infection studies, using a variety of plant-VAM fungus combinations, have clearly
demonstrated that VA myconhiza development is well defined and highly predictable
(see Harley and Smith, 1983). Two types of plant root cell can be affected; the
epidermal and outer cortical cells within which simple hyphal coils may develop and
the inner cortical cells which usually contain the highly branched, haustorium-like
arbuscules.
Ultrastructural and cytochemical analyses indicate that the fungal coils are surrounded
by a thick layer of continuous and frbrillar material (interfacial matrix) which appears
to be unmodified primary cell wall (Bonfante-Fasolo, 1988). The cytoplasm, central
vacuole and amyloplasts appear normal (Bonfante-Fasolo, 1987). Cells containing
arbuscules show distinct cytological modifications; the invading hypha penetrates the
cell wall and invaginates the plasmalemma which substantially proliferates (Cox and
Sanders, 1974; Toth and Miller, 1984) around the rapidly branching hyphae to
accomodate the increased surface area and a thin layer of interfacial matrix is
synthesised between the fungal cell wall and the plant plasmalemma (Scannerini and
Bonfante-Fasoslo, 197 9).
V/hen compared to non-mycorrhizal inner cortical cells, the newly synthesised
plasmalemma and cytoplasm adjacent to the interfacial matrix show some distinct
modifications. Jabaji-HaÍe et aI. (199O) used lectin-gold complexes to demonstrate that
N-acetyl-D-glucosamine, N-acetyl-D-galactosamine, D-mannose and sialic acid sugar
residues were more abundant in arbuscular cells of Glomus clarum -infected leeks,
particularly in the cytoplasm and the plasmalemma surrounding the plant-fungus
interface. Increased neutral phosphatase activity, frequently associated with cell wall
synthesis (Jeanmaire et a1.,1985) and ATPase activity (Ma¡x et a1.,1982; Gianinazzi-
Pearson et aI., 1991) has been localised (using Pb deposition) to this region of the
plant plasmalemma. Physiological and biochemical studies strongly suggest that
membrane embedded proton-pumping (H+) ATPases control many aspects of plant
t7
physiology (see Serrano, 1989; Sussman and Harper, 1989) including solute transport
coupled to the pH and electrochemical gradients generated by these enzymes (Serrano,
1939). There are some legitimate critisisms of the cytochemical procedures used to
detect H+-ATPase activity (see Katz et aI., 1988; Smith and Smith, 1990; Gianinazzi-
Pearson et aI., l99l). However, it seems fair to conclude that these and other
modifications which are not observable around the hyphal coils or in uninfected cells,
indicate specialised alterations to the host membrane sufrounding the fine a¡buscule
branches (Smith and Giananazzi-Pearson, 1988) which are consistent with the
argument that this is the site of bidirectional nutrient transport.
In a comparative study of nuclei from thç roots of uninfected leek and leek infected by
a Glomus sp., Berta et aI. (1990) presented evidence that the nuclei from the
mycorrhizal roots were significantly hypertrophied. The authors suggested that this
was probably due to chromatin decondensation, an ¿ugument which they supported
with transmission electron micrographs. They went on to speculate that these results
may be related to increased transcriptional activity as a result of the Glomts infection.
As yet there is no quantitative data to support this hypothesis but two other lines of
evidence do suggest modifications to plant-gene expression: the cytoplasm adjacent to
the interfacial matrix appears to contain much higher concentrations of ribonucleic acid
(RNA) than the cytoplasm of non-myconhizal cortical cells (Berta et a1.,1990) and the
profiles of soluble proteins extracted from mycorrhizal roots of onion, tobacco
(Dumas, et aI., l9S9) and soybean (Pakovsky, 1989) show qualitative and/or
quantitative changes in comparison to non-mycorrhizal roots. It is almost certainly true
that some of the new proteins detected in these studies are a¡tifactual anÜor of fungal
origin (Guttenberger and Hampp, t992) or represent plant defence-related proteins
(see Harrison and Dixon, 1994). Cytochemical localisation of proteins within
developing mycorrhizas, using antibodies raised to these proteins may be the only way
of unambiguously determining their origin. The purification of a putative acid chitinase
18
which is present, independent of the fungus involved in the symbiosis, is under way
(see Dumas-Gaudot et a1.,1992).
Incompatibitity in Plant-VAM Fungus Interactions1.8
Whilst the overwhelming majority of plant species are susceptible to infection by VAM
fungi, a number of plant genera do not appear to form typical or normal VA
mycorrhizas. Unfortunately, much of the data collected on non-mycorrhizal or poorly-
mycorrhizal plant genera has been obtained from field grown plants. As Tester ¿r a/.
(1987) have pointed out, there could be several explanations for the appatent lack of
infection in these samples; the plant roots may not have come into contact with viable
(or compatible) inoculum; infection may be seasonal and the samples were collected at
the wrong times; the local soil environment may not be conducive to, or may actively
discourage infection (ie. high soil P, waterlogging, residual pesticide/herbicide
activity); or the plants may be inherently non-mycorrhizal.
There is also the added problem of defining mycorrhizal status in order to score
infection. Some VAM fungi-plant combinations do not form vesicles but do form
functional mycorrhizas while others can lack normal arbuscules or appear as extensive
but atypical infections (Harley and Smith, 1983; McGee, 1985) which may or may not
be functionally active. Glasshouse experiments using controlled inoculum levels co-
ordinated with detailed infection studies are absolutely necessary to accurately
determine whether a particular species/genus/family of plants is capable of forming
functional mycorrhizas (for a complete summary of these problems see Tester et aI.,
1987).
Never-the-less it has been established that several plant families can be considered
genuinely and generally non-mycorrhizal under controlled glasshouse conditions. Of
these the Brassicaceae and the Chenopodaceae are the best characterised (see Hirrel er
al., 1978). Other important families include the Proteaceae (Khan 1978; Lamont
1982), the Zygophyllaceae (Khan t974) and the Restionaceae (Lamont 1982). Several
19
non-mycolrhizal genera. most notably Lupinus (Morely and Mosse 1976; Trinick
1977), occur in families which are considered to be mycorrhizal.
How non-mychorrhizal plants resist or greatly restrict infection by VAM fungi is not
known, but this incompatibility suggests that these species are fundamentally different
from the majority of vascular plants. A substantial amount of work has concentrated
on identifying fungitoxic compounds which accumulate within or are exuded from the
roots of non-mycorrhizal plant species (see Koide and Schreiner,1992).
Glucosinolate hydrolysis products, such as isothyocyanates (mustard oil) have
received the most attention but the results are inconclusive as there does not appear to
be a strong inverse relationship between root concentrations of these compounds and
infection. Isothyocyanate-producing species (such as Carica papaya ) which would be
expected to display restricted infection, form normal mycorrhizas (Peterson et aI.,
1985). Brassica sp. which display marked phenotypic variation in root glucosinolate
concentration (7-524mmoVgf.wt) remain uninfectible (Glennl et al., 1985) while
normal infections were produced in compatible hosts grown in close proximity to these
plants, indicating that root exudates are probably not a determinant of VA mycorrhiza
incompatibility in Brassica sp. either (Glenn, et a1.,1988). In contrast, root exudates
from Lupinus albus do appear to. hinder hyphal attachment and appressorium
formation (Giovannettt et al.,1993) and intergenic grafts between Lupinus and Pisum
can induce resistance to VA mycorrhizal symbiosis in the pea roots, suggesting the
pÍesence of inhibitory systemic shoot factors in Lupinus sp. (Gianinazzi-Pea¡son and
Gianinazzi, 1992).
Infection in non-host plants can sometimes be induced by the close proximity of an
infected mycorrhizal host plant (see Hinel et a1.,1978). In these cicumstances the
infected host plant may provide compounds (nutritional or otherwise) critical in
stimulating hyphal growth and development which allows the subsequent infection of
a non-host plant (Tester et al., 1887). These taxa may prove to be particularly
important in elucidating the mechanisms which determine the outcome of VAM
20
fungus-plant interactions. There is very little known about the molecular or genetic
basis of mycorrhizal development. The incompatibility observed in certain plant genera
may reflect the absence or mutation of a particular plant gene product or the evolution
of specific resistance mechanisms. The latter appears more likely given the apparent
diversity of resistance stategies.
Recently, alfalfa (Medicago sativa L.), pea (Pisum sotivum L.) and faba bean (Vicia
faba..L.) genotypes defective in mycorrhiza formation have been reported (Bradbury
et aI., l99l;Duc et aI., 1989). The true value and significance of these particular
genotypes remains unclear because the classical genetic studies they undertook indicate
that these mutations are tightly linked or involve recessive mutations to (as yet
uncharacterised) Rhizobium symbiosis loci which clearly complicates further
molecular or genetic characterisation of these particular plant genotypes: My
interpretation of these data, and those presented by Gollotte et al. (1993) does not rule
out the induction of a general root resistance-like response in these genotypes and in
any event, the analysis of a biochemical pathway involved in the establishment of both
the Rhizobium a¡d VA mycorrhizal symbioses (both of which are often associated
with dramatic nufiitional effects), quickly becomes complicated. However, as Duc ¿t
aI. (1989) point out, these plants could prove to be extremely useful for investigating
genetic variabilty and specificity in VA myconhizal fungi.
rWhat is urgently required is a set of plant genotypes, wild-type or induced, deficient at
various stages of mycorrhizal development; ideally the altered phenotypes should
range from very early infection stages (such as fungal attachment and appressorium
development) through to the later stages (arbuscule and vesicle development).
Preferably they should be derived from a non-leguminous plant species or cultiva¡
which has been genetically well characterised so as to allow a comparative analysis of
the two genotypes (see Smith, 1995).
2t
1.9 Aims and Significance of this Project
There is good evidence that host plant cells are cytologically modified in direct
response to infection by VA mycorrhizal fr¡ngi. It is reasonable to suggest that this is a
result of altered gene expression (possibly by de novo transcriptional activation) within
host plant cells.
The overall aim of this project was to develop a model system suitable for undertaking
a molecula¡ analysis of mycorrhiza development, and to determine if particular host
plant genes are expressed during the early stages of infection. The types of questions
to which answers are sought include:
(Ð Is there an association between the early stages of mycorrhiza
development and the accumulation of specific mRNA species?
(ü) How do these mRNA species accumulate temporally and spacially
within the mycorrhizal plant root?
(üi) What are the functions of the mRNA Fanslation products?
This project was divided into 3 distinct stages:
(Ð The development of a model VAM fungus/plant system suitable
for molecular analysis.
(ü) The production of a mutagenised plant population and subsequent
identification of plant mutants deficient in mycorrhiza formation.
(üi) Cloning and cha¡acterisation of putative mycorrhiza-related mRNA
species.
This research is significant for several reasons. It has the potential to identify some of
the molecular events which occur in this association and would represent the first
reports of this nature. The information obtained may ultimately be used to identify the
biochemical events which control the symbiosis. It is possible that the mycorrhizal
association represents a "basic compatibility" (Ellingboe, 1981) upon which the action
22
of specific resistance genes have been superimposed in planUpathogen symbioses.
Understanding the molecular basis of compatibility may help in elucidating how plants
respond during incompatible interactions. Finally, the development of a well
characterised model system and the isolation of plant genotypes deficient in mycorrhiza
development and/or function could prove to be extremely valuable for future
researchers wishing to carry on this work.
23
Chaoter 2
Materials and Methods
2.1 Plant and Fungal Genetic Stocks
2.1.1 Plants
Seeds of Hordeum vulgare L. cv. 'Galleon' and cv. 'Schooner' were supplied by Dr.
R.G. Lance, Dept. of Plant Science. Barley addition and ditelomeric lines were
provided by Dr. Ken Shepherd, Dept. of Plant Science. The doubled haploid mapping
populations were prepared by Drs Sue Logue (Chebec x Harrington and Galleon
Hanrna Nijo) and Rafrque Islam (Clþer x Sahara).
2.1.2 Fungi
Glomus intraradices Schenk and Smith inoculum was obtained from pot cultures of
TriþIium subterraneum L. grown in a soil:sand mix as described below. All pot
cultures were supplied by Dr. S.E. Smittr.
Gaeumannomyces graminis var. tritici (Ggt-800) which was isolated from Triticum
aestivum at Avon, S.A. by A.D. Rovira and G. graminis var. graminis (Ggg-W2P)
which was isolated from a Paspalidium sp. at rWalcha, N.S.\ry. by P.T.W. V/ong
were obtained from Dr. Paul Harvey. These were supplied as pure cultures gro$/n on
PDA medium.
24
2.2 Plant Growth Conditions
2.2.1 Potting Medium
Plants were grown in a mixture of steamed sand and autoclaved soil (9:1 w/w). Plants
were grown in either white plastic pots supplied by Disposable Plastic Products or
containers made from black plastic bags. The pot and container sizes varied between
experiments (see the relevant chapters) and neither contained drain holes,
2.2.2 Seed Sterilisation
Seeds were washed in NaOCI (O.SVo v/v) for 2 minutes and then washed 2-3 times rn
autoclaved HZO (the H2O used in all of the investigations described herein was of
nanopure quality). This was followed by 2x2 minute washes in 90Vo ethanol and a
further 3 washes with autoclaved HZO.The seeds were then placed in autoclaved
glass Petri dishes containing 3 sheets of sterilised filterpaper and 15ml of sterilised
IJZO. The Petri dishes were covered in aluminium foil and left at22-25oC for 36-48
hours.
2.3 Plant Growth
The growth conditions were either in a glasshouse without artificial light when the
natural daylength was less than 11 hours or in a growth cabinet with 11 hours of
artificial light (600-700mE). The day and night temperatures were 18-20oC and l4oC
respectively.
All plants received mineral nutrients (-P unless otherwise stated) at the beginning of
the growth period and at 2 weekly intervals. These nutrients were made from stock
solutions (see Appendix 1) when required and applied at the rate of lOml/kg of
sand:soil. In some experiments (see Chapters 3 and 6) a range of phosphate (P) levels
were used.
25
The following P levels apply where indicated:
Ps :No added phosphate
P1 :Phosphate added at the rate of 0.5mmoles/kg of soil:sand mix.
P2 :Phosphate added at the rate of l.5mmoles/kg of soil:sand mix.
Peø :No added phosphate at the beginning of the growth period but
phosphate was added at the rate of 0.Smmoles/kg of soil after
12 days growth.
2.4 Harvesting of Plant Material
Plants were harvested at the times indicated in each experiment.The roots were quickly
but carefully washed in distilled water to minimise tissue damage and the accumulation
of mRNA transcripts which might have been induced by the harvesting process. The
roots were then blotted dry with tissue paper. Plant growth of roots and shoots was
evaluated by fresh weight (immediately after careful blotting) or dry weight (after
incubation at 80oC overnight). Root and shoot material required for nucleic acid
extraction was snap frozen in liquid nirogen within 2 minutes of harvesting.
2.5 Assessment of Mycorrhizal Colonisation
The extent of mycorrhizal infection in the roots was assessed as follows: Root samples
were washed until free of the potting medium, cut into lOmm pieces and cleared by
incubation tn a LOVo KOH solution for 4 hours at 55oC. The cleared root sub-samples
were stained with Trypan Btue (Phillips and Hayman, L97O) and the extent of
mycorrhizal colonisation (as a percentage of total root length) of each root sub-sample
was determined using a grid intersect method (Giovannetti and Mosse, 1980).
Mycorrhizal root sections were usually scored on a 1-3 scale which described the type
of colonisation: (1) intercellular infection only, (2) arbuscules present and (3)
arbuscules and vesicles present.
26
2.6 Other Materials.
A large number of enzymes, bacterial strains, media and solutions \ilere used in the
course of this work. The vast majority are those which are commonly used for
investigations requiring recombinant DNA technology and have been described in
Maniatis et at. (7983). I have listed those that I used and their composition (where
applicable) in Appendix 1.
2.7 General Recombinant DNA Methods
2.7.1 Escherichía Coli Methods
2.7.t.1 Bacterial Cultures
Starter cultures and small scale plasmid extraction cultures were prepared by
inoculating 10ml of culture medium (LB or TFB) containing the appropriate antibiotic
with a single colony of. E. coli from a fresh agar plate. They were then incubated 6-8
hours at37oC with shaking.
Large scale cell cultures (50-250ml) were prepared by inoculating culnue medium (LB
or TFB) containing the appropriate antibiotic with L-znl of fresh starter culture and
incubating as described above until the desired optical density of the culture was
reached.
2.7.1.2 Preparation of Escherichía coli Competent Cells
LB (l50ml) in a 250m1 centrifuge flask was inoculated with the appropriate E. coli
strain starter culture (lml) and incubated at 37oC with shaking until the A699 was
approximately 0.6. The cells were then cooled in iced water before being centrifuged
(5000rpm/5 minutest{oc).The supernatant was poured off and cold 50mM
CaCl2llQmM Tris (pH 8.0) solution (75m1) was added. The cells were gently
resuspended and and left on ice for 15 minutes before being centrifuged (l000rpm/5
minuty'ethanol precipitation ( c,ells were then very gently resuspended in cold 50mMI
CaCI2 (5rnl) whilst swirling the flasks in iced water. The cells were either used for the
transformation immediately or divided into 200p1 aliquots and prepared for snap
27
Frfpezing in liquid nitrogen and storage at -75oC after the addition of glycerol
(30ttUaliquot)
2.7.1.3 Escheríchia colí Transformations
Fresh competent cells or gently thawed frozen competent cells (200p1) were dispensed
into Eppendorf tubes. Plasmid DNA (5¡rl) was added and the solution gently mixed.
The celUDNA mixtr¡re was then left on ice for 30 minutes before being heat shocked in
a 42oC water bath for 2 minutes. The mixtrue was incubated on ice for 3 minutes and
then at room temperature for 5 minuæs. LB (800p1) was added and the cells incubated
for a further 2 hours before spreading on LB l.S%o agar plates with the appropriate
antibiotic to select for plasmid transformation and then incubated overnight at 37oC. A
single colony was picked and cultured overnight in LB medium plus antibiotic
(5O¡rg/ml ampicillin when selecting for pTZlSU).
2.7.2 RNA Methods
2.7.2.1 Isolation of Plant Total RNA
The mortars, pestles, spatulas and glassware used for RNA isolations were oven-
baked at 160oC for a minimum of 6 hours to inactivate exogenous RNase activity.
Ha¡vested root or shoot material (0.5-l.0gm) was snap-frozen in liquid nitrogen,
placed in a morta¡ and ground to a fine powder. More tiquid nitrogen was added to
pool the root material at the bottom of the mortar before grinding for another 60
seconds. The frozen powder was transferred to a 30ml Corgx centrifuge tubel.-5*.1 (toci^tlT.rs \+ctret{g,rt:lO'" c4 ÉõTA r teolø-þc Uo"ï1).
containing(nf Uuffer. Th'e slüny was vortexed (30 seconds) before
centrifugation (11000 rpm/10 miri/4oC) in a lA20 rotor. The supernatant was
transferred to a 10ml Beckman ultracentrifuge tube containing 1g of CsCl added for
each lml of supernatant. Three ml of a CsCl solution (O.9759hnl) was laid into the
bottom of the tube using a sterilised Pasteur pipette. The tube was then centrifuged
(30000 rpmll4-L6 hours/6oC) in a Ti 70.1 rotor to pellet the total RNA.
28
After centrifugation, the sticky surface layer was removed using sterilised cotton buds
and the supernatant removed with a baked Pasteur pipette. The sides of the tube were
carefully and thoroughly wiped to remove as much of the residual supernatant as
possible before the pellet was resuspended in 400¡tt of cold RE buffer (this was done
on ice to minimise residual RNAse activity) and transferred to a 1.5m1 Eppendorf tube
containing phenoVchloroform/isoamylalcohol (400p1). After gentle vortexing (30
seconds) and centrifugation (l2000rpm/5 minutes) the upper phase was transferred to
a fresh tube and ethanol-precipitated with the addition of 3M sodium acetate (a0pl) and
lml of cold ethanol. The total RNA pellet was washed with 70Vo ethanol (lml) before
resuspension in an appropriate volume of RNase-free HZO.
2.7.2.2 Synthesis of Double-stranded cDNA and Cloning into ì,gt10
Double-stranded cDNA with linked Eco Rl/Not I adaptors was prepared using a
cDNA synthesis kit purchased from Promega. The manufacturers protocols were
followed precisely. The cDNA product was isopropanol precipitated and resuspended
in lOpl of nanopure HZO.
The linkered cDNA (2pl) was ligated for 12 hours at 15oC with EcoR I
digested/dephosphorylated ?r.gt10 arms (0.5pg) in a final volume of 5pl. The ligation
mix was then ethanol precipitated and washed with 807o ethanol prior to packaging
and transfection of competant cells.
2.7.2.3 Packaging and Transfection of Phage Library
Packaging of the cDNA/tr gt10 ligation mix into infective bacteriophage À particles was
carried out according to the manufacturers (Promega) instructions. The packaging-
reaction mixes were diluted to 500p1 with SM buffer and 15pl of chloroforrn was
added to prevent contaminating bacteria from growing.
29
The Eansfection of packaged DNA into E coli strains C600 añC6ffihfl was carried
out as follows:
A starter culture was prepared by inoculating LB (50m1) containing 10mM
MgSO¿.and O.27o maltose, with an appropriate E. coli strain and incubating the
culnue overnight at37oC with shaking.
The recipient cells were prepared during the DNA packaging incubation period by
inoculating LB medium (20ml) containing 10mM MgSO4.and0.2Vo maltose, with the
overnight culture (500p1). The culture was incubated at 37oC with shaking until the
optical density at 600nm was approximately 0.6 units. This incubation was ca¡ried out
in 50ml flasks to ensure optimal aeration. The cells were then transfered to a 25ml
centrifuge tube and pelletted (300Orpm/l0minutesl4oc) in the SS34 rotor. The
supernatant was discarded and the cells resuspended in cold 10 mM MgSO¿ (lOml).
An aliquot of the cell suspension (100-300p1) was then inoculated with packaged
DNA (100p1), gently mixed and incubated at25oC for 15 minutes with no shaking.
Molten warm Top Agarose (4-l2ml depending on the plate size) containing 10mM
MgSOa was added and the mixture gently vortexed before being overlayed onto a
room temperature LB agar plate. The plate was incubated for 8-12 hours at37oC.
2.7.2.4 RNA Gel Electrophoresis
All RNA separations were performed in a mini-gel apparatus under RNA denaturing
conditions using a l.2%o agarose MOPSÆDTA gel matrix and loading buffer
containing formaldehyde (2.2M) and formamide (5O7o). The RNA (5 or l0pg/lane)
was dried under vacuum and resuspended in RNA load buffer (4.5p1) and
formamide/formaldehyde (9.5¡rl). The sample was incubated at 70oC for 10 minutes
to denature the RNA secondary structures and then chilled on ice for 2 minutes. Gel
loading buffer (3pl) was added and well mixed before loading into the gel. Samples
were separated at 3OmA for 30 minutes and then 60mA for a further l-2 hours.
30
2.7.2.5 RNA Gel Blotting, Prehybridisation and llybridisation
RNA species (total or poly A+ preparations) were separated by RNA-denaturing
electrophoresis and transfered to Hybond Nr membrane by capillary blotting in
l0xSSC. The membranes were fixed by baking at 80oC for 2 hours. Prehybridisation
and hybridisation procedures were essentially the same as for DNA hybridisations
except that RNA prehybrisation and RNA hybridisation solutions (see appendix 1)
were used.
2.7.2.6 Synthesis of Radiolabelled First-strand cDNA
Total RNA (10-30lrg) or mRNA (0.1-0.5¡rg) was made up to a final volume of l0¡rl
using RNase-free HZO.One ¡rl of oligo (dT) tZ-tg (500mg/ml) was added and the
mixture incubated at 70oC for l0 minutes before quick chilling on ice. First strand
reaction mixture (see appendix 1) was prepared (8trl) and added to the RNA/oligo-
nucleotide mixture. M-MLV H- reverse transcriptase (200 units) was then added and
the reaction mix incubated at 37oC for 45-60 minutes. After completion of the
reaction, a lpl sample was removed to a new Eppendorf tube and mixed with 9pl of
TE buffer. Then lpl of this solution was spotted onto one DE50 filter and 5pl onto a
second filter. The incorporation of a-32P-dCTP into the newly synthesised DNA was
checked by the trichloroacetic acid precipitation of the DNA and comparison of the
relative Cherenkov radiation signals (Maniatis et al.1982). First strand cDNA probes"ri rp8.p-l-qwith a specific acti. ivity in excess offere purifred using a Sephadex G-100 column.
2.7.3 DNA Methods
2.7.3.1 DNA Isolations from Plant and Fungal Tissues
(Ð Root and fungal tissue
High molecular weight DNA (suitable for PCR analysis or restriction endonuclease
digestion) was isolated from small amounts (10-100mg) of root tissue and fungal
spores using a modification of the method descibed by Saghai-Ma¡oof et aI., (1984).
31
The procedure, which uses cetyttrimethylammonium bromide (CTAB) buffer, is
described at Appendix 2.
(ü) Iæaf tissue
The isolation of high molecular weight plant DNA from leaf tissue was canied out as
described at Appendix 3.
2.7.3.2 Plasmid Isolation from Bscheríchia coli Transformants
(Ð Small scale isolation of plasmid DNA
Plasmids for immediate restriction-digest analysis were prep¿¡red by a modification of
the alkaline lysis method (Birnboim and Doly, L979). The protocol is described in
Appendix 4. Plasmid DNA for sequencing was also prepared as described in
Appendix 4 but the final pellet was further purifred by PEG precipitation.
(ü) Large scale isolations of plasmid DNA
Plasmids for purification and storage were prepared from larger volumes (250 or
500m1) of cell cultures as described by the Maniatis et al. (1982).
2.7.3.3 Purifïcation of Large-Scale Plasmid DNA Preparations
Large-scale preparations of plasmid DNA were purified on caesium chloride-ethidium
bromide density gradients. The method was as described by Maniatis et al. (1982)
except that only 600¡rl of ethidium bromide was added, no light parafin oil was used
and the Beckman ultralock tubes were centrifuged at 50000rpm in the Ti-70.1 rotor for
only 14-16 hours. The bands were visualised and extracted under long-wave ultra-
violet light. Ethidium bromide removal, dialysis and DNA precipitation and
resuspension were as described by Maniatis et aI. (1982).
32
2.7.3.4 Isolation of Bacteriophage DNA
Bacteriophage larnbda DNA was isolated using a modification of method described by
Amersham in the cDNA cloning-kit manual and uses DEAE-52 cellulose to remove
contaminating E. coli DNA from the lysate. The method is described in Appendix 5.
2.7.3.5 DNA Precipitations
(Ð Ethanol precipiøtions
One tenth volume 3M NaAcetate (pH 4.5) was added to help all precipitations, except
those being carried out in high salt buffers (ç. precipitation of minipreparation
plasmid DNA). Two volumes of cold absolute ethanol were then added and the
solution left to stand for at least 30 minutes at -20oC. The DNA was pelleted by
centrifugation, washed with 70Vo ethanol to remove salts, vacuum dried, and
resuspended in the desired volume of nanopure H2O or TE buffer.
(ü) Isopropanolprecipitations
One tenth volume of 3M KAcetate (pH a.Ð was added to all precipitation reactions.
One volume of isopropanol was then added and the solution left to stand at room
temperature for 5-10 minutes. The DNA was pelleted by centrifugation, washed twice
with cold TOVo ethanol, vacuum dried and resuspended in the desired volume of
nanopure H2O or TE buffer.
(üi) Polyethylene glycolprecipitations
Preparations of DNA to be used for sequencing reactions included an extra
precipitation using polyethylene glycol (PEG). Ethanol or isopropanol precipitated
DNA was resuspended in 32pt of nanopureHZO. 4M NaCl (8¡rl) was mixed with the
DNA sample before the addition of l3%o PEG (40td).The solution was then incubated
on ice for 30-60 minutes. The DNA was pelleted by centrifugation at 4oC, washed
33
twice with cold TOVo ethanol, briefly vacuum dried and resuspended in l0-20p1 of
nanopure HZO.
2.7.3.6 DNA Gel ElectroPhoresis
DNA fragment separations were carried out by submerged, horizontal gel
electrophoresis in TBE agarose or TAE agarose gels in their respective conductive
buffer solutions. The TAE agarose gels were used only for qualitative restriction-
digest fragment analysis or for fragment purifications using the 'Gene-clean' system.
Rapid sreening of restriction digests or DNA preparations was ca¡ried out using mini-
gel apparatus. Overnight electrophoresis was ca¡ried out in large tanks. The DNA
fragments separated by gel electrophoresis were stained with ethidium bromide (l
-/rytnd) to enable visualisation or photography of the bands under short or long-wave
lengthUVlight.
2.7.3.7 DNA Eel Blotting, Prehybridisation and Hybridisation
(Ð Plaque hybridisation studies
Recombinant 1,gt10 DNA was denatured and fîxed to Hybond N+ using a procedure
simila¡ to the bacterial colony transfer method described by Benton and Davis (1978):
'lPhage were plated out as previously described using 13.5cm disposable Petri dishes.
A dry, circular, l3cm diameter Hybond Nr membrane filter was placed on the surface
of the bacterial lawn containing the phage plaques and marked so as to permit
unambiguous determination of the orientation of the membrane relative to the plate at a
later time. The membrane was then carefully removed and briefly air dried, colony
side up, before being treated, in assembly line fashion, by exposure to the following
solutions: (1) denaturing (0.5M NaOlVl.5M NaCl) for 3-5 minutes, (2) neutralisation
(0.5M Tris-HCl pH 7.5fl.5M NaCl) for 5-10 minutes and (3) 2x SSC for l-2
minutes. These treatments were carried out in trays containing fwo sheets of Whatman
3 MM filter paper saturated in the appropriate solution. Membranes were treated
34
colony side up and never submerged as this usually caused streaking. The aim was to
treat the transferred phage DNA by exposure to the solutions through the membrane.
After air drying on a single sheet of Whatman 3MM paper, the filter was placed in a
'Hybaid' hybridisation bottle and prehybridised in 20ml of DNA prehybridisation
solution (see appendix 1) for 4-8 hours. The DNA prehybridisation solution was then
replaced with lOml of freshly prepared DNA hybridisation solution (see appendix 1)
plus denatured radioactive probe (200pt/bottle) and salmon sperm DNA (l00pg/ml
final concentration). The membrane was left to hybridise with the probe for 18 hours
at 65oC with constant rotation in the Hybaid oven. The filters were then removed
from the boffles and agitated in 200m1 of wash buffers 1 and then 2 (see appendix 1)
at 68oC for 25 minutes to remove excess probe and hybridisation solution. The
washings were repeated twice before the filter papers were exposed to X-ray film at
-80oC for a minimum of 48 hours using intensifying screens.
(ü) Restrictiondigested-DNAhybridisationstudies
DNA fragments were separated by O.7-2Vo agrirose TEB gel electrophoresis and then
transferred to Hybond N+ membrane using the capillary blotting method (Southern,
t975). The membrane was placed in a Hybaid hybridisation bottle containing 20ml of
DNA prehybridisation solution (see appendix 1) and prehybridised for 4-12 hours at
65oC in a Hybaid hybridisation oven with constant rotation.
The DNA prehybridisation solution was replaced with DNA hybridisation solution
(l$ml), denatured radioactive probe (200m1) and salmon sperm DNA (lO0ug/ml final
concentration), the hybridisation was then allowed to proceed for 12-18 hours at 65oC
in the oven with constant rotation. The membrane was removed and washed twice
with constant agitation in wash buffer I (200my65o920 minutes) and once in wash
buffer 2 (2O0mU65oCl2O minutes) to remove the hybridisation solution and excess
probe. Membranes which showed considerable background radioactivity after these
35
treatments were w¿rshed in wash buffer 3 (2OOrrIl65oC/10 minutes) The membranes
were then exposed to X-ray film at -80oC using intensifying screens.
2.7.3.8 Restriction Endonuclease Digests
Complete restriction digests were carried out in buffer conditions recom{nended by the- __tiq*au p€' dt\arlrån)
enzyme supplier. Sufficient restriction enzymeþas added to compleíely digest the
DNA in 2 hours at37oC.
2.7.3.9 Isolation of DNA Fragments from Electrophoresis Gels
The required DNA fragments were cut out of an agarose gel under long-wave UV
light and the DNA purified using'Gene-clean'as described by the manufacturer @IO
101). The size and approximate DNA concentration of the excised fragments were
then checked by electophoretic separation and ethidium bromide visualisation against
known standard concentrations.
2.7.3.10 Ligations
Ligation of restriction-fragment DNA into designated restriction sites of phosphatase-
treated plasmid vector pTZl8U and cDNA into the Eco Rl restriction site of the
bacteriophage vector l,gtl0 was ca¡ried out as described by Maniatis et aI. (1982).
Phosphatase treaÍnents were also carried out as described by Maniatis et aI. (1982).
2.7.11 DNA Radio-labelling for Hybridisation Studies
AII probes for hybridisation studies were prepared by random primer radio-labelling
of double stranded DNA with cr-32P-dCTP. The DNA template (80-100ng) plus
randomlmer primer mix (3¡rl) was boiled for 5 minutes and then chilled on ice for 2
minutes. This was then added to a fresh Eppendorf tube containing 2x reaction buffer
(L2.5,f) anda-32P-dCTP (1-3pl). The reaction mixture was made up to a volume of
24¡tl with nanopure H2O before the addition of I unit of Klenow enzyme (lu/pl).
Finally, the sample was incubated at 37oC for 30-60 minutes.
36
2.7.3.12 Purification of Radio-labelled DNA
A sterilised pasteur pipette was loaded with Sephadex G-100 in TE buffer and washed
with TE buffer containing 0.17o SDS (1-2ml). The labelling reaction mixture then was
made up to a final volume of 200¡rl with TE buffer,loaded onto the column and eluted
with TE buffer in 200p1 aliquots. The radioactivity of each aliquot was measured
using a hand-held Geiger-Mueller counter and nvo peaks were always recorded for the
elution curve. The first peak represented a-32p-dCTP incorporated into the
synthesised DNA, and the second unincorpor ated a-32P-dCTP.
2.7.3.13 DNA Sequencing using Tøq DNA Polymerase
Recombinant plasmids for sequencing were isolated and PEG precipiøted as described
above. The DNA concentration of each plasmid preparation was determined by gel
electrophoresis and adjusted to approximately 500ng/Ff . A2¡tl aliquot was tansferred
to a fresh Eppendorf tube containing 9pl of nanopure HZO. The diluted plasmid
sample was used for the subsequent sequencing reactions. These were carried out
using a modified dideoxy method (Sanger et aI., 1977) with fluorescent dye linked
primers, buffer and Taq DNA polymerase supplied by Apptied Biosystems The
reaction products were combined, ethanol precipitated and briefly dried. Separation of
the sequencing reaction products was carried out on a Applied Biosystem automatic
sequencer by either Dr. N Shirley or Ms. J. Nield. The sequence data was analysed
using SeqEd software (Applied Biosystems) running on a Macintosh Centris 650
personal computer.
2.7.4 Polymerase Chain Reaction (PCR) Methods
2.7.4.1 PCR Amplification of 1,gt10 Inserts
The bacteriophage vector ÀgtlO contains a unique EcoRI site within the phage 434
immunity region (imm434). Cloning into this site allows selection for recombinant
phage. The use of primers, specific for and complementary to the irnm434 regions on
3',7
either side of this cloning site enabled reliable PCR amplification of DNA inserts up to
3 kilobase pahs.
The phage plaque was cored out using a sterile Pasteur pipette and transferred into an
1.5m1 Eppendorf tube containing SM buffer (50-100p1). The tube was incubated at
4oC for a minimum of 2 hours to allow diffusion of the phage particles into the
buffer. A 10pl aliquot of the phage sample was then transferred to a small PCR
Eppendorf tube and boiled for 10 minutes to denature the phage protein coat and the
DNA. The boiled sample was chilled on ice and spun down before the addition of the
PCR buffer mix (15 or 40ml) which contained 20 mM Tris-HCl (pH8.4), 50 mM
KCl, 1.25 mM MgCl, 250 mM dNTPs, phage-F and phage-R primers (0.1 pg of
each; see appendix 7 for all the primer sequences used in this worþ and 0.5 units of
Taq DNA polymerase. The reactions were ca¡ried out in an MJ-Research Thermal
Cycler programmed as follows: denaturation of DNA at 95oC for 4 minutes followed
by 32 or 36 cyles of I minute denaturation at 94oC, 2 minutes primer annealing at
55oC and2 minutes DNA chain extension at74oC. The chain extension for the final
cycle was allowed to proceed for 7 minutes before the tubes were cooled to 4oC.
2.7.4.2 PCR of Plant Genomic DNA
Total plant genomic DNA (0.5 pg) was diluted 1:100 in SM buffer (see appendix 1).
An aliquot (10ttl) was transferred to a small PCR Eppendorf tube containing 40pl of
PCR reaction buffer containing 20 mM Tris-HCl (pH8.4), 50 mM KCl, 1.75 mM
MgCl, 250 mM dNTPs, PCR primers (0.1 pg of each) and 0.5 units of Taq DNA
polymerase. The reactions were carried out in an MJ-Research Thermal Cycler
programmed as follows: denaturation of DNA at 95oC for 4 minutes followed by 36
cyles of 1 minute denaturation at 94oC, 2 minutes primer annealing at 45oC and2
minutes DNA chain extension at74oC. The chain extension for the final cycle was
allowed to proceed for 10 minutes before the tubes were cooled to 4oC.
38
Pla,smid or phage insert amplifications for qualiøtive analyses and production of DNA
for probes were ca¡ried out using the same conditions as described above except that
the thermal cycler was prograrnmed as follows: Denaturation of DNA at 95oC for 3
minutes followed by 36 cyles of I minute denaturation at 94oC, I minute primer
annealing at 55oC and 1 minute 30 seconds DNA chain extension at74oC. The chain
extension for the final cycle was allowed to proceed for 10 minutes before the tubes
were cooled to 4oC.
39
r
Growth Response and Infection Studies ofthe Hordeum vulgare - Glomus intraradices
Myc orrhizal Symbio sis.
3.1 General Introduction
Vesicular-arbuscular (VA) myconhizal resea¡ch has in the main cented on whole plant
or gross cellular responses to plant colonisation by VA mycorrhiza fungi. A relatively
small number of host plants have been used for much of this research; chosen because
they are mycorrhizally responsive, easy to grow and have large root cells which make
clearing, staining, sectioning and visualisation of anatomical stn¡ctures comparatively
straightforward. Spring barley (Hordeum vulgare L.) is an economically important
cereal species, widely grown in South Australia and meets all the above criteria as well
several others which are briefly discussed below.
This investigation of the mycorrhizal symbiosis will use recombinant DNA technology
and ionising radiation mutagenesis and will require several additional features in the
host plant; it must be amenable to nucleic acid extraction; it must be genetically well
characterised, diploid and self-fertilising; and ideally, it should also be genetically
transformable and re-generable from ex planta material.It is important that the extent
of colonisation can be manipulated to ensurs a large number of early-infection units are
present in the tissue being used for the proposed analyses. This should mærimise the
chances of cloning and detecting low abundance mRNA transcripts involved in the
establishment and mainænance of VA mycorrhizas.
40
The rwaite Institute has a successful barley breeding program which has concentrated
on cyto- and molecular genetic approaches in recent ye¿ìrs. As a result chrosomomal
mapping of cloned barley genes is relatively simple; barley cv. 'Betzes' addition and
ditelosomic lines in a'Chinese Spring' wheat background are available (Islam et aI.,
l98l) and the research group of Dr. P. Langridge has considerable experience in
cereal nucleic acid manipulations and is well advanced in the preparation of a genetic
map of barley using doubled haploid populations generated from several crosses.
Barley fransformation and ex planta regeneration a¡e not yet routine techniques but two
research groups at the Waite Institute are working towards these ends (Drs. A. Aryan
and S. Logue respectively). This is an extremely important (although long term)
consideration because the final, definitive test of a genes putative function will involve
the transformation (complementation) of a suitable (usually mutant) phenotype to
restore the putative function of that genes translation product to that particular plant
phenotype. Finally, barley remains an important experimental organism for plant
mutagenesis experimentation and considerable expertise is available in this through Dr.
R. Lance.
Glomus intraradic¿s Schenk and Smith, was chosen as the VA fungus for the
experiments involving mycorrhizas described in this work because it readily colonises
a broad range of host species (including barley) and forms morphologically and
functionally 'normal' mycorrhizas (S. Smith pers. comm'; see Chapter 1).
Differential growth responses to VA mycorrhizal colonisation have been repofed
among cultivars of clover (Crush and Caradus, 1980), corn (Hall, 1978), cowpea
(Rajapakse and Miller, 1987), pea (Estaun et aI., 1987), pearl millet (Krishna et al.,
1985), tomato (Bryla and Koide, 1990), soybean (Skipper and Smith, 1979) and
wheat (Azcon and Ocampo, 1981; Manske, 1989). Genotypic variation also appears to
influence the extent of root colonisation by VA mycorrhizal fungi (Azcon and
Ocampo, 1981; Toth et aI., L984; Krishna et a1.,1985; Young et aI., 1985; Estaun ¿f
aI., 1987; Heckman and Angle, 1987; Rajapakse and Miller, 1988; Manske, 1989;
4l
Sreenivasa and Rajashekhara, 1989; Boyetchko and Tewari, 1990; Bryla and Koide,
1990; Mercy et a1.,1990; Toth ¿f a1.,1990; Vierheilig and Ocampo, 1991). There a¡e
relatively few reports on VA mycorrhizal colonisation and growth responses in and
between cultivars of Hordeum spp.However, Baon et aI. (1992 and 1993) who
worked concurrently with work discussed in this thesis, showed clear differences in
the extent of mycorrhizal colonisation in a range of South Austalian barley cultivars.
It has not been possible to always correlate the extent of colonisation with the variable
growth responses observed because most of the studies described above have failed to
take total root length into consideration. Accordingly, the variable responses described
by these authors may reflect differences in extent of colonisation, mycorrhizal nutrient
aquisition effrciency or phosphate utilisation efficiency. These serve to emphasise that
the choice of host plant genotype is an important consideration and that any given plant
genotype/fungus genotype combination needs to be evaluated, bearing the aims of the
proposed investigation in mind, to ma¡rimise the chances of a sucessful outcome.
Field nutrition trials have indicated that'Galleon', a South Australian barley cultivar,
is more responsive (in terms of certain growth parameters) to phosphate amendments
than many other South Australian barley cultivars, including 'Schooner' (Rob
Wheeler, personal comm.). The growth and nutritional effects of VA mycorrhiza
development are well established and have been extensively reviewed (Smith, 1980;
Cooper, 1985; Hayman, 1982; Smith and Gianinazzi-Pearson, 1988). Increased
growth in response to myconhiza development in soils with low available P
concentrations indicates the formation of a functioning symbiosis; ie. the roots have
become mycorrhizal and nutrient exchange has occurred.
The formation of a functioning or effective mycorrhiza also infers that the plant has
responded 'normally' to colonisation by the VA mycorrhizal fungus in terms of the
expression of genes which may facilitate and/or control VA mycorrhiza development
or mediate nutrient transfer. A large growth response by either of these barley cultivars
(relative to the other) could indicate greater differences in the expression of genes
42
necessary for effective mycorrhiza development and maintenance. This could reflect
greater susceptibility to colonisation (and the total number of plants cells involved in
the symbiosis) and/or greater changes in the up- or down-regulation of symbiosis-
related gene expression within individual cells. Mærimising host plant responses to
VA mycorrhizal fungl colonisation within a given amount of tissue was an extremely
important aim of this project because the experimental approach proposed for cloning
prospective symbiosis-related genes has a greater chance of success if enriched tissues
a¡e used for the cDNA library preparation and subsequent differential screening.
The colonisation studies were also an important preliminary to the production and
screening of a mutagenised barley population for mycorrhiza-deficient phenotypes (see
Chapter 4). Jensen and Nittler (1971) reported striking varietal differences in leaf- and
sheath-colour between and among barley seedlings grown with complete nutrient
solution or a nutrient solution lacking in phosphorus. Leaf- and sheath-colour were
documented during the experiment in torder to determine whether barley seedlings
grown in low P soils without G.intraradices displayed P deficiency symptoms ea¡lier
and/or more severely than those grown with the VA mycorrhiza fungus. A simple
system for nutritionally screening a mutant barley population for phenotypes which do
not form effective mycorrhizas obviates the need to check each root system
individuallyi the advantages in time and resources are clear.
The general aims of the two experiments reported in this chapter were to:
(Ð confirm the mycorrhizal status of barley.
(ü) choose a cultiva¡ suitable for these experiments.
(üi) establish a time course for colonisation of the roots which could be
used to routinely produce mycorrhizal tissue of a predictable
developmental stage for molecular genetic analyses.
(iv) determine whether phosphate deficiency could be visually detected
in young seedlings and whether early mycorrhizal colonisation
negated this effect.
43
3.2 Experiment 1: Variation between two
Hordeum vulgare Cultivars in Response to
Glomus intrarødíces Colonisation.
3.2.1 General Introduction
This first experiment compared the growth responses and extent of. Glomus
intraradices Schenck and Smith colonisationn22 and 36 day old plants of the barley
cultivars 'Galleon' and 'Schooner'. The specific aim was to ascertain if one or both of
these cultivars were suitable host-plants for molecular-genetic analysis of the
interaction.
The extent of root colonisation by the VA mycorrhizal fungus G. intraradic¿s was
measured at each of the harvests. The aim was to collect infection during this
experiment to confirm that the plants became mycorrhizal under the growth conditions
used and to provide information on the extent and physiological stage of VA
mycorrhiza colonisation within the roots of each cultiva¡ at certain chronological ages.
Because the rates of VA mycorrhiza colonisation in developing barley roots are Poorly
documented, this information was necessary in order to design an colonisation-time
course experiment to investigate more fully the development of.G. intraradices induced
mycorrhizas in the roots of the selected host plant.
3.2.2 Experimental Methods
There were two ha¡vests (21 and 36 days after transplantation) and six Eeatments for
each cultivar: the plants were grown with or without G. intraradices (M+ or NM) at
three phosphate levels (Po, Pt and Pz). Each treatment was replicated three times for a
total of 72 plants.
One hundred and fifty seeds (seventy five per cultivar) of approximately the same size
were selected for germination. The seeds were surface sterilised, germinated and
planted out after 36 hours (42for each cultivar) into black plastic bags (one seed/bag)
4
containing 1.5kg of the soil:sand growth medium (see 2.2). The growth medium for
the M+ treatments contained G.intaradices pot inoculum (lOVo w:w) whilst the
phosphate treatments contained either no added phosphate (Po); phosphate added at
0.5 mmoles/kg (P1¡; or phosphate added at 1.5 mmoles/kg (Pz). Liquid Phosphate
Solution was used for the phosphate amendments and Nutrient Solutions A-E were
added at the rate of l0ml/kg of growth medium (see Appendix l). All the nutrients
were then watered into the soil. The plants were kept at field capacity (120m1,/kg soil)
by daily watering to weight.
The barley plants were grou/n in a glasshouse under natural light. The daylength was
less than 11 hours and the day/night temperatures ranged from 18-24oC and 10-l5oc
respectively. These plants were visually assessed for phosphate deficiency 10, 12,14,
16, 18, 20,25 and 30 days after transplantation. Destmctive harvests were ca¡ried out
at 22 days and 36 days. At each ha¡vest, the shoots lvere cut off at the soil level and
weighed before drying at 75oC. The roots were washed until free of the potting
medium, cut off below the remains of the seed, surface dried and weighed. A small
sample (20-40mg) was excised from each root system and weighed before being
clea¡ed by incubation in a KOH solution (see 2.5). The remainder of each root system
was dried at75oC and weighed. The fresh-weight:dry-weight ratios of the residual
root portions were used to calculate the dry weight of each whole root system. Three
days later the cleared root sub-samples were used to determine the extent of
mycorrhizal colonisation of each root sub-sample was determined using a grid
intersect method (see 2.5).
3.2.3 Results and Discussion
All of the germinated seeds survived the transfer procedure and all plants inoculated
with G. intraradices became mycorrhizal. The extent of colonisation as a percentage of
total root length for each treatment afe presented in Figures 3.1(a) and 3.1(b).
Colonisation percentages were higher in 'Galleon' than 'Schooner' at both harvests
and for all treatments. The extent of colonisation of the M+:P1 and M+:P2 plants
45
(both cultivars) was reduced compared to the M+:Pg plants. Increased soil
concentrations of P have frequently been shown to result in reduced colonisation (see
Mosse, L973: Abbott et al., 1984; Amijee et aI., 1989; Koide and Li, 1990). The
mechanism is unknown (see Koide, 1991) but it appears that plant-tissue P
concentration may have a regulatory role in the development of colonisation (Sanders,
1975; Menge et aI., 1978 and Azcon et al., 1978). There were no significant
differences in the proportions of the va¡ious colonisation stages present in the roots of
'Schooner' or 'Galleon' (data for M+Pg plants is shown at Figure 3.2). Stage (2) and
(3) mycorrhizas (see 2.5) totalled approximately 65Vo and 85Vo of the mycorrhizal
tissues of each cultiva¡ (for any of the phosphate treatments) at harvests I and 2,
respectively.
After l0 days growth the seedlings of both cultivars appeared to be uniform with
respect to height and had normal (dark green) leaf-sheath colour. Discoloration of the
leaf-sheath became noticeable n25%o of all Pg and Pr plants after 15 days growth. By
the first harvest 77Vo plants had some degree of leaf-sheath discoloration and some
33Vo of.the first and second leaf tips were chlorotic. It was not possible to ldentify the
low phosphorous or mycorrhizal treatments or the cultivar by monitoring the onset of
P deficiency symptoms, it was apparcnt that early mycorrhizal colonisation had no
effect on the leaf-sheaf discoloration caused by P deficiency in the barley cultivars
'Galleon' and'Schooner'.
The NM:P6 plants of both cultivars grew poorly. Total root and shoot fresh and dry
weights and root artd shoot fresh weight /dry weight ratios were lower relative to all
other treatments and root/shoot ratios were higher than for all other treatments (Table
3.1). These data are indicative of severe P deficiencies (see Harley and Smith, 1983)'
The shoots and root systems of 'Schooner' were smaller (dry weight) than 'Galleon'
for all treatments. VA mycorrhizas stimulated the shoot and root growth of all P9
plants but had no significant effect on plants which received additional P. The M+:Pg
plants of 'Galleon' 'trere more responsive to VA mycorrhiza formation with G.
46
intraradices than those of 'Schooner'. The M+:Pg 'Galleon' plants obtained 22 day
old fresh and dry root/shoot weights which were much closer to those achieved by
plants grown with additional P.
After the first harvest, the NM:P6 plants of both cultivars grew very poorly in
comparison with plants in other treatments. The NM:Pg plants were much shorter and
their leaf-sheath colour remained purple/red. Many first and second leaves became
completely chlorotic and eventually necrotic. In stark contast, most of the other plants
(including all the'Galleon'M+:Pg plants) regained their normal sheath colouration by
day 30. VA mycorrhiza formation was clearly correlated with alleviation of the P
deficiency symptoms. There were three exceptions: two individual pots from the
'Schooner' NM:P1 treatments and one from the'Galleon'M+:Po: treatment. By the
time of the second harvest (36 days) nutrient deficiency symptoms were again visible
in 83Vo of the plants (the plants a¡e shown in Figure 3.3). All plants harvested at day
36 had large root sytems which filled the pot and had become matted at the bottom of
the pots; it is quite likely that the accessable soil P or other nutrient reserves had been
exhausted. The growth data collected at the second ha¡vest paralleled those obtained
from the first (see Table 3.2). However, these data should be considered less useful
than those obtained from the first harvest because of inter-root competition and
exhaustion of nutrients.
These results show that under any of the growth conditions used for this experiment,
'Galleon' plants had larger root and shoot sytems and higher levels of VA mycorrhiza
colonisation than 'Schooner'. 'Galleon' was also more responsive than 'Schooner' (in
terms of growth) to VA mycorrhiza formation in the Pg $owth medium. Whether this
response represents more efficient aquisition of soil nutrients by the mycorrhizal
'Galleon' roots or better utilisation of the aquired nutrients by the 'Galleon' plant is
unclear and would require further analyses (see Smith et a1.,19921' Baon et a1.,1992).
V/hat is clear, is that after only 22 days'Galleon' formed effective mycorrhizas and
more of the root system was infected than'Schooner'.
47
The barley cultivar 'Galleon' was selected as the host plant for the remainder of this
work because it best met the criteria discussed in the introduction to this experiment:
'Galleon' was responsive to mycorrhiza colonisation and 'Galleon' roots appeared to
be a more enriched source of mycorrhizal tissues. Mycorrhizal, low P or adequate P
status was not distinguishable by monitoring the onset of nutrient difficiency
symptoms in either cultivar and so it was concluded that nutritional screening of the
planned mutant barley population for the presence or absence of functional
mycorrhizas would not be possible.
The levels of colonisation at the 22 day harvest showed that showed that good levels
of colonisation (537o) could be achieved in 'Galleon' within 22 days but that most of
the colonisation units contained arbuscules and/or vesicles. I decided to concentrate
initially on genes involved in the early stages of colonisation (see Chapter 1) primarily
because the symbiosis becomes more complex as the mycorrhizas develop. The
mixture of new and physiologically older (including senescing) mycorrhizal tissue is
likely to involve more complex and overlapping patterns of plant gene expression.
Concentrating on the early stages also minimises the total amount of fungal tissue
relative to the plant tissue and so reduces the chances of cloning fungal cDNA's.
These problems were best overcome by working with mycorrhizal barley root tissues
that contained as many early-infection units as possible, associated with minimal
development of arbuscules and vesicules. It was clear from Experiment 1 that the
optimal stage and time for harvesting root material was before 22 days. The next
experiment was designed to investigate the time course of Glomus intraradices
colonisation in 'Galleon' prior to22 days growth.
FIGURE 3.1
Extent ofGlomus intraradices colonisation (as a percentage of total root length) in
22 day old (3.1a) and 36 day old (3.1b) Hordeum vulgare cv. 'Galleon' and
'Schooner' plants grown at three levels of added phosphate; no added phosphate
(P0); phosphate added at 0.5 mmoles/kg (P1); and phosphate added at 1.5 mmoleslkg
(P2). Samples from three plants for each treatment were assessed for total
colonisation and extent of colonisation containing developing arbuscules. The
standard errors of the means a¡e shown.
Figure 3.1a
60
Figure 3.lb
50
g[tirtiìTruNñ
Galleon total col.Galleon arbs.Schooner total col.Schooner arbs.
50ñÉ.9 40
6úU)
s30I+r:20oxr¡l 10
0
PO P1
Phosphate treatment
PI
P2
ioÐ40oCË.2 30ooO
b20c.)
XEl 10
0
PO
Phosphate treatment
P2
FIGURE 3.2
Proportion of mycorrhizal tissue containing arbuscules in22 day old (Harvest 1) and
36 day old (Harvest 2) Hordeum vulgare cv. 'Galleon' and 'Schooner' roots
colonised by Glomus intraradices and grown without additional phosphate (M+PO).
Samples from three plants for each harvest were assessed for total colonisation and
extent of colonisation containing developing arbuscules. The extent of mycorrhizas
containing arbuscules is expressed as a proportion of the total colonisation. The
standard errors of the means are shown.
Harvest 1
E GalleonSchooner
Harvest 2
Figure 3.2
100
80
60
40
20
EEõoÉE!l cst+ ct)
ooåEa.¡ Ë=OL)e3uåt
0
TABLE 3.1
Effects of Glomus intraradices colonisation and phosphate on shoot and root growth
and shoot and root physiology of 22 day old Hordeum vulgare cv. 'Galleon'
seedlingg. The means and standa¡d errors of the means (S.E.M.) of the three
replicates of each treatment are shown. The treatments were nonmycorhizal (NM),
mycorrhizal (M+), no added phosphate (PO), phosphate added at the rate of 0.5
mmoles/kg (P1) and phosphate added at the rate of 1.5 mmoles/kg.
GAIIEON
Growth characteristic
Shoot fresh weight (g)
Shoot FAM S.E.M.Shoot dry weight (g)
Shoot D/TV S.E.M.
Shoot fresh/dry weight ratio
Root fresh weight (g)Root F/lV S.E.M.Root dry weight (g)Root D/TV S.E.M.
Root fresh/dry-weight ratio
Root/shoot ratio (f.w.)
SCHOONER
Growth characteristic
Shoot fresh weight (gms)
Shoot FÆV S.E.M.Shoot dry weight (gms)Shoot D/lV S.E.M.
Root fresh weght (gms)Root FÆV S.E.M.Root dry weight (gms)Root DÆV S.E.M.
Root fresh/dry-weight ratio
Root/shoot ratio (f.w.)
Treatment::P0 NM:PI NM:P2
0.69 1.610.03 0.140.11 0.230.01 0.01
r.720.160.240.02
r.32o.t20.19o.02
+
r.670.060.100.02
0.9 30.040.1 I0.01
M+:P21.7 5
0.t20.240.02
6.4t 6.93 7.18 7.06 7.10 7.22
0.520.050.080.01
0.840.090.1 I0.01
0.910.070.1 I0.01
0.940.030.110.01
0.770.020.100.01
6.45 7.87 8.15 7.61 8.22 8.53
0.75 0.52 0.s3 0.58 0.s6 0.s4
TreatmentNM0.78 t.r40.07 0.040.08 0.160.01 0.01
1.190.090. 15
0.01
0.9 3
0.rl0.100.02
1.200.030.140.01
M+:PO M+:Pl M+:P2r.t20.070.140.01
Shoot fresh/dry weight ratio 10.29 7.08 8.22 9.78 8.02 8.61
0.600.010.050.01
0.680.060.1 I0.01
0.540.060.060.01
0.6 30.1 1
0.1 I0.01
0.670.060.rI0.01
0.640.040.100.01
12.57 6.31 6.t6 9.2't 5.66 6.t9
0.78 0.s6 0.56 0.s8 0.s6 0.56
TABLE 3.2
Effects of Glomtx íntraradices colonisation and phosphate on shoot and root growth
and shoot and root physiology of 36 day old Hordeurn vulgare cv. 'Galleon'
seedlings. The means and standa¡d errors of the means (S.E.M.) of the three
replicates of each treaünent are shown. The treatments were nonmycorrhizal (NM),
mycorrhizal (M+), no added phosphate (PO), phosphate added at the rate of 0.5
mmoles/kg (P1) and phosphate added at the rate of 1.5 mmoles/kg.
Shoot fresh/dry weight ratio 5 .39 6.39 6. 13 5 .96 5.8 3 5.7 6
GALLEON
Growth characteristic
Shoot fresh weight (g)Shoot F/TV S.E.M.Shoot dry weight (g)Shoot FAM S.E.M.
Root fresh weght (g)Root F/TV S.E.M.Root dry weight (g)Root D/TV S.E.M.
Root/shoot ratio (F. wgt.)
SCHOONER
Growth characteristic:
Shoot fresh weight (g)Shoot FAA¡ S.E.M.Shoot dry weight (g)Shoot DÆV S.E.M.
Root fresh weight (g)Root FÆV S.E.M.Root dry weight (g)Root D/TV S.E.M.
Root fresh/dry weight ratio
Root/shoot ratio (F. wgt.)
Treatment.NM1.82 7 .400.19 0.r70.34 1.160.01 0.06
Treatment:NM:P0 NM:PI
7.220.071.360.09
0.870.050.150.01
7.51o.2tt.230.06
4.7 8
o.230.800.05
7 .r40.08r.240.09
0.620.040.100.00
1.030.060. 15
0.01
0.810.020.r20.01
0.940.030.140.01
t.t40.090.r70.01
I NM:P2 M+:PO M+:Pl7 .33o.r21.260.07
r.100.050.r70.02
+ M+:Pl M+:P27 .300.231.550.06
0.7 4
0.040. 13
0.01
Root fresh/dry weight ratio 6.03 6.90 6.63 6.62 6.85 6.78
0.34 0. 14 0. 15 0. 17 0. 13 0. 16
2.390.07o.5z0.02
7 .380. r91.360.1 I
2.8 8
0.08o.670.0s
7 .010.6 3
1.510.08
Shoot fresh/dry weight ratio 4 .60 5.29 5.43 4.29 4 .65 4.70
0.550.030.100.00
0.840.030. 14
0.01
0.470.060.080.01
o.'t90.010.130.00
5.74 5.89 6.17 6.27 s.74 6.07
0.23 0.r2 0.11 0.16 0.11 0.11
FIGURE 3.3
The effect of VA mycorrhizal colonisation on shoot morphology and growth of
Hordeum vulgare cv. 'Galleon' seedlings. The seedlings were grown for 35 days
with (M+) or without (NM) Glomus intraradices and with (P+) or without (P -) of
added phosphate (0.5 mmoles/kg).
M+ P+ M+P- : NM P+ NM P-
9ì
53
3.3 Experiment 2 - Rate and Extent of VA
Mycorrhiza Development in Glomus intra¡adices
Colonised Hordeum vulgare cv 'Galleon' Roots
3.3.1 Introduction
The major objective of this project was to identify plant genes which mediate the
development of the symbiosis so the experimental procedure adopted needed to be
reproducible to ensure that I could obtain plant root material containing mycorrhizas of
a reasonably predictable physiological age. The fust experiment had been carried out
in a glasshouse which va¡ied in light intensity daily and seasonally. Light intensity can
affect mycorrhiza development (Son and Smith, 1988) but does not appear to have an
effect on spore germination or early colonisation events. On this basis a growth
cabinet was used for this and all subsequent experiments involving the production of
mycorrhizal 'Galleon' root tissue. The plants were gtown in smaller (400 gm) plastic
pots to minimise use of inoculum, the sand:soil growth medium and growth room
space requirements. There is no evidence to suggest that a smaller volume of growth
medium would dramatically alter the colonisation ontogeny during the early stages of
VA mycorrhizaformation. The G. intraradices inoculum density was raised from
lOVo to líVo (wlw) in an attempt to increase the number of early colonisation units.
Non-mycorrhizal control plants, with and without added P, were grown and compared
with the M+ plants to ensure that 'Galleon' rvas still responsive to mycorrhiza
colonisation (indicating the formation of a functional mycorrhiza) under the slightly
altered growth conditions.
54
In this experiment, colonisation of the developing root systems of 'Galleon' seedlings
by G.intrarad.ic¿s was followed more closely. The aims were:
(1)
(2)
to develop a reproducible system for generating mycorrhizal
'Galleon' seedlings, predominantly at early stages of
colonisation, and
to identify the chronological age at which these roots should be
han¡ested to meet the objectives discussed in the introduction
to this experiment.
3.3.2 Experimental Methods
This experiment measured the extent of mycorrhizal colonisation as it developed over
23 days in newly germinated barley cv. 'Galleon' seedlings planted out into potting
medium containing G. intraradices, r,,
One hundred and forty 'Galleon' seeds were surface sterilised, germinated and
transplanted (see 2.2) into individual pots containing 400gm of soil:sand (1:9 w/w)
potting medium and G. intraradíces inoculum (l5Vo wlw) (M+:PO). A further 40
seeds were transplanted into individual pots containing 400gm of the soil:sand potting
medium only (NM:Pg) or 400gm of potting medium with P added at the rate of 0.5
mmoleslkg (NM:Pr). Nutrient solutions were added and watered into the soil as
previously described. The plants were grown in a controlled growth cabinet as and
kept at field capacity (48 ml,/kg soil) by regular watering to weight.
Five M+:Pg, three NM:P9 and three NM:P1 seedlings were randomly selected for
harvesting every second day from the day I to day 23 after transplantation. The shoot
and root fresh weights were recorded before the extent of mycorrhizal colonisation
within the roots was determined (2.5).
55
3.3.3 Results and Discussion
Slight leaf chlorosis and leaf sheath discoloration was noticable in 3 of the plants after
15 days growth and became evident in all (surviving) plants by day 21. The plants
achieved slightly lower final fresh weights at day 25 (see Table 3.3) than those
recorded at day 22 inBxperiment 1. This could have been due to lower irradiance
levels. The irradiance in the growth cabinet was not measured at the time at the time of
the experiment but later measurements (A. Bruce pers. comm.) showed that it was
much lower than in the glasshouse. Root/shoot ratios were higher in NM:P9 control
plants than in either the M+:Pg or NM:P1 plants, indicating physiological changes
resulting from inadequate P availability. The growth response data indicates similar
responses to the experimental conditions in this experiment as in the comparative
study.
None of the NM control plants and all of the M+ plants developed normal mycorrhizas
(see Figure 3.4). The first colonisations were not detected until 7 days after
transplanøtion. Lag times before initiation of colonisation have been well documented
(Saif and Khan, 1977; Carting et al., 1979) and at least partially reflect the time
required for infection-propagule germination, hyphal elongation and root penetration
(Sutton and Barron,1972: Sutton, L973).
The extent of colonisation increased with the age of the plants to a mærimum mean
value of nearly 58Vo of the root length at day 21 before decreasing to 52 Vo at the end
of the experiment. This decrease is not statistically signifrcant but may reflect increased
rate of growth by the developing root system. Figure 3.5 illustrates the ontogeny of
VA mycorrhizal development over the course of this experiment. Mycorrhizal tissues
with developing arbuscules were detected as early as day 9 but remained at less than
l7o of the total root length scored (or lOVo of the total mycorrhizal tissue) until day 13;
the final value was 42Vo of the total root length (or 8l7o of the total mycorrhizal
tissue).
56
Whilst scoring the roots for colonisation, it became apparent that very few (and in
some samples none), of the thicker primary roots of 'Galleon' contained G.
intraradices infection units. The lag time befween transplantation and the detection of
mycorrhizal infection unis could therefore also reflect the growth pattern of the barley
root system. The large primary roots clearly represent a large proportion of the root
system during the first few days of growth and development. If they are only poorly
susceptible to mycorrhizal formation, colonisation may not be possible until secondary
roots are produced a few days after transplantation. This also suggests that VA
mycorrhizas are not evenly distributed within barley root systems and that harvesting
only secondary roots could provide an even more enriched supply of 'Galleon' cells
involved in the interaction with G. intraradices .
'Galleon' and G. intraradices readily form functional mycorrhizal tissues under the
growth conditions used. As Figure 3.5 clearly shows, strong early-stage colonisation
(12-377o of the total root length) containing low proportions of arbuscules (less than
lOVo) cønbe obtained in'Galleon'after 11-13 days growth under these experimental
conditions. The low proportion of infection units observed in the primary roots
required confirmation in view of the stated aim of using root tissue which contains the
highest proportion of early stage colonisation for this study. This observation was
followed up in Experiment 3.3.
TABLE 3.3
Increase in shoot and root fresh weights of Hordeumvulgare cv. 'Galleon' seedlings
g¡own wittr (M+) and without (NM) the VA mycorrhizal fungus Glomus intradices.
The plants were hanested and weighed every two days for 23 days, starting one day
after transplantation into the growth medium. The means of each treaunent (three
replicates for the NM plants and five replicates for the M+ plants) and the standa¡d
errors of those means (S.E.M.) are shown.
Table 3.3
Day of harvest135 79111315r7192123
Root fresh weight (g)
NM MeanS.E.M.
MeanS.E.M.
NM MeanS.E.M.
M+ MeanS.E.M.
M+
0.030.01
0.040.01
0.090.01
0.090.01
0.090.01
0.090.01
0.140.01
0.120.01
0.160.01
0.230.03
0.330.01
0.390.06
0.380.01
0.420.01
0.480.05
0.640.08
0.620.06
0.5r0.01
0.550.04
0.660.08
0.730.06
1.260.06
0.720.02
0.610.04
o.t70.01
0.24o.o2
0.260.01
0.290.03
0.31o.02
0.470.04
0.880.05
0.620.01
0.530.02
1.11
0.0s
0.550.03
0.030.01
0.040.01
0. 13
0.01
0.140.01
0.190.01
0.2r0.01
0.390.02
0.390.04
0.M0.03
0.460.02
0.490.01
0.480.02
0.69o.o2
FIGURE 3.4
Ontogeny of VA mycorrhiza developmentin Hordeum vulgare cv. 'Galleon' roots.
The roots were harvested after 7,9, tl and 21 days growth then cleared and stained
as described at2.5. The roots were photographed through a light microscope at 40X
magnification.
Ar = arbuscule; Ap = appressorium; IH = intercellular hyphae; S = spore and V =
vesicle.
7 DAYS
DAYS
9 DAYS
2I DAYS
Ir
.l :' ì. .. (,
S. Ar
V
^Ap
la
FIGURE 3.5
Graph showing the time course of colonisation in Hordeum vulgare cv. 'Galleon'
seedlings inoculated with Glomus intraradic¿s.. Five replicate plants were harvested
every two days, starting one day after transplantation into the soil medium, then
cleared, stained and assessed for total colonisation and extent of colonisation
containing developing arbuscules as described (2.5). The means and standard errors
of those means are shown.
Figure 3.5
60
50
40
30
20
10
----o Total Colonisation---{- - Alb. Coionisation
o
U)
ao
(-)
F
0
0 .5 10 15
Day of Harvest
20 25
60
3.4
3.4.1
Experiment 3 - The Distribution of
Glomas intraradices Colonisation in the Roots
of Hordeum vulgare cY. tGalleont.
Introduction
This experiment was undertaken to test:
(l) the observation that VA mycorhhizas developed primarily on the
lateral roots of 'Galleon' barley, and
(2) the reproducibility of the results obtained in Experiment 2.
3.4.2 Experimental Methods
Forty 'Galleon' seeds of approximately the same size were surface sterilised,
germinated and transplanted into individual pots containing 400 gm of soil:sand (1:9
w/w) potting medium and G. intraradices inoculum (LSVo w/w) as previously
described (2.2).4 further 7 seeds were transplanted into 400 gm of the sand:soil
medium only. Nutrient solutions A-E were added at the rate of 10 ml/kg of growth
medium and watered in. The pots were placed in the growth room and kept at field
capacity by watering to weight (,+48 gm) every third day.
After 13 days, the shoots and intact root systems of 5 randomly selected plants from
each treatment were carefully harvested and weighed. The root systems were cleared
with l07o KOH and stained with Trypan Blue (see 2.5) prior to being laid out on a
sheet of glass and immersed in a glycerol:H20 solution (1:l v/v) to imobilise the roots
and stop them from drying out. The length of the individual primary roots was
determined using the graph paper grid and the number and length of each individual
lateral root on each primary root and the number of the VA mycorrhizal infection units
on each of those lateral roots were recorded. The extent of colonisation, as a
percentage of the total length was also assessed using the graph paper grid placed
under the glass plate. The VA mycorrhizal infection units were clearly discernable by
6l
eye. However, a dissecting miscoscope was used to check for the presence of very
early stages of colonisation (attachment and appressorial swellings only). The
infection unit data were expressed as a percentage of the root system using the total
root length data.
3.4.3 Results and Discussion
The root and shoot fresh weight and root length data for the 'Galleon' plants (see
Figures 3.6(a) and 3.6O)) indicate that the NM and M+ plants were aPproximately the
same size at the time of harvesting. No nutrient deficiency symptoms lvere obvious;
atl plants maintained a deep green sheath and shoot colour. The development, shoot
and root size (fresh weight) of these plants were very similar to those of the plants
harvested at day 13 in Experiment 2.
The colonisation data for the M+ plants is shown in Figure 3.7. The mean extent of
colonisation of the whole root system for the 5 plants was 42Vo, which is higher than
the value of 37Vo reported in 13 day old 'Galleon' in Experiment 2. This does not
appear to be significant. The grid intersect method (Giovannetti and Mosse, 1980)
tends to overestimate actual root colonisation (McGonigle et a1.,1990; Bradbury er
aI., t99l) but may also underestimate it if the colonisation is unevenly distributed and
a sub-sample is used to quantify VA mycorrhizal colonisation. It is clear from this
experimental data that the extent of G. intraradices colonisation was unevenly
distributed within the root system. It was highest in the lateral roots and in particular
the first 5 laterals of each primary root.
The proportion of infection units with arbuscules was highest in the primary roots
(where extent of colonisation was the lowest). It was also higher in the oldest five
lateral roots in comparison with the whole lateral root system. The colonisation data
from Experiment 2 suggests that the first infections occur at around 7 days. At least
some of the VA mycorrhizal colonisation in the flrst five laterals (and the primary
roots) of Experiment 3 could therefore be up to 8 days old. Arbuscules can develop
62
within this period of time (see Harley and Smith, 1983; Toth and Miller, 1984) and
so, are likely to occur in greater proportions in the regions of older mycorrhizal tissue.
FIGURE 3.6
Shoot and root growth data (3.6a) and root length data (3.6b) for 14 day old
Hordeum vulgare cv. 'Galleon' seedlings grown with (M+) and without (NM) the
VA mycorrhizal fungus Glomus intraradices. Five replicates of each treatment were
harvested after 14 days growth. The shoots and roots were weighed and the primary,
lateral and total root lengths determined by immobilising the entire root systems in a
large petrie dish which was placed over 1mm graph paper. The mean of each set of
measurements and the standard errors of those means are shown.
0.5
0.4
0.3
bo
ÉÒo
c)>ØC)li
tJr
Figure 3.6 (a)
0.7
0.6
0.2
0.1
Figure 3.6 (b)
I Non-mycorrhizal
Myconhizal
Shoot
Tissue type
E Non-mycorrhizal
Mycorrhizal
Lateral
Root tissue
0
Root
2000
1600
1200
800
400
b0
o
ú
0
Primary Total
FIGURE 3.7
Distribution and extent of Glomus intraradices colonisation in 13 day old Hordeum
vulgare cv. 'Galleon' roots. The intact root systems of five plants were harvested,
cleared and stained as described (2.5). The intact root systems were then spread out
in large petri dishes and placed over lmm graph paper. The total colonisation and
extent of colonisation containing developing arbuscules within the primary roots, the
lateral roots, the first five lateral roots and the entire root system and distribution of
colonisation was then determined. The means of these data and the standard errors of
those means a¡e shown.
Figure 3.7
80
70
60
50
40
30
20
10
0
Total infectionArb. infection
ño(-)c)(lr
+r
o)IXr!
Primary Laterals Laterals 1-5
Root tissue
Total
6s
3.5 Summary and Conclusions
Experiment 1 established that the two f/. vulgare L. cultivars, 'Galleon' and
'Schooner', were susceptible to colonisation by G. intraradices and formed effective
VA mycorrhizas in the low P soil:sand growth medium: The levels of colonisation
obtained in 'Galleon' were higher than in 'Schooner'.
Experiment I also showed that 'Galleon' was more responsive to VA mycorrhiza
formation in terms of shoot and root growth. 'Galleon' was selected as the plant host
for the remainder of this investigation because these results may also indicate greater
responses in terms of the overall expression (up and down regulation) of symbiosis-
related plant genes.
Experiment 2 showed that the optimal period for harvesting'Galleon'root material for
the proposed molecular anlyses was 11-15 days after transplanting freshly germinated
seeds into the soil:sand growth medium inoculated with G. intraradices Pot inoculum
(l5%o w/w). During this period the proportion of VA mycorrhizal tissue containing
a¡buscules was low and the extent of colonisation was between 22-37Vo. At this stage
of mycorrhiza development, the plants showed no significant differences in root and
shoot growth.
Higher levels of colonisation could be obtained by harvesting later, but the proportion
mycorrhizal tissue containing arbuscules quickly increased to over TOVo by day 19.
Leaf and sheath discoloration and chlorosis indicated P deficiency in many of the
plants from day 19 onwards. The shoot and root growth data supported this
conclusion. Unfortunately, early mycorrhiza development has no effect on the onset
of these deficiency symptoms and so nutritional screening for mycorrhiza-deficient
phenotypes is not possible.
Experiment 3 confirmed that the colonisation method used in Experiment 2 was
reproducible. It also clearly showed that the distribution of VA mycorrhizal
colonisation within the 'Galleon' root system 13 days after transplantation into the
66
inoculated soil was not uniform. The great majority of the colonisation was
concentrated in the lateral roots. The first 5 lateral roots proved to be a source of
highly-infected mycorrhizal root tissue but probably contained too many arbuscules
for the preparation of a cDNA library to be used for the isolation of genes involved in
the early stages of symbiosis establishment. The top half of M+ root systems (where
colonisation is concentrated) were selected for the library prepartion as this appeared
to be a reasonable compromise between using whole roots and further
experimentation to optimise the level and type of colonisation in the oldest lateral
roots.
67
Isolation of Putative Mycorrhiza-
deficient (myc -) Barley Phenotypes
4.1 General Introduction
Mutations are a source of variability in organisms. The availability of well
characterised mutants provide valuable biological material for detailed molecula¡
studies because complex metabolic pathways and developmental processes can be
analysed using mutants which a¡e defective at well defined and identifrable stages (see
Cove, 1993).
Mutations occur spontaneously but can also be induced by a variety of physical and
chemical agents (mutagens). Induction of mutation is a useful method for obtaining
mutant genotypes for two reasons; the frequency of identifiable mutation events is high
(up to several orders of magnitude greater than for a spontaneous mutation) and the
basic genotype of the plant is usually only slightly altered. The generation of near
isogenic lines in which many (if not most) attributes remain unaltered allows direct
comparisons of mutant genotypes with a well cha¡acterised parental plant. Mutants
which do display unwanted phenotypic characteristics can be relatively easily
backcrossed to the nearly isogenic parent genotype and the progeny re-screened for
plants with the desired mutation(s) but free of the unwanted trait(s).
A range of mutant plant genotypes, suitable for physiological, biochemical and
molecular studies have been generated in several species (see Koorneef, 1991). Recent
examples include cesium-insensitive, storage-protein accumulation and flower
development mutants inArabidopsis thaliana (Sheahan et al., 1993; Namba¡a et al.,
68
1992; Bowman et aI., 1989), seed coat colour mutants in beans (Guimaraes et aI.
1990), proanthocyanidin-free barley (Jende-Strid and Moller, 1981; see Jende-Strid,
1991), mitochondrial DNA mutants in tobacco (Bonnett et aI., 1993), and
pigmentation mutations tn Antirrhimum (Luo et aI., l99l).
Mutants have been particularly useful in the analysis of bacterial (Escherichia coli ,
Agrobacterium tumefaciens and Rhizobium spp.) and yeast (Saccharomyces
cereviseae) gene-enzyme and developmental pathways (for recent examples see Evans
et a1.,1993; Mantis and rù[inans, 1993; Roth and Stacey, 1989; Westenberg et al.,
1989; Folch et aI., 1989). The relatively well characterised legume-Rhizobium
nitrogen fixing symbioses serve best to highlight both the potential and difficulty of
plant mutagenesis experimentation. The primary focus of symbiotic nitrogen fixation
research has been on the microbial parmer and considerable progress has been made in
identifying key molecular and biochemical events nRhizobiurt spp. which lead to an
effective association between the legume and the bacterial partner (see Dénaré et al.,
1990; Long, 1989a and 1989b; and Sprent, 1989). This is in no small way due to the
relative ease of producing, identifying and isolating a range of suitable bacterial
mutantü
Any plant species is inherently more difficult to mutagenise, manipulate and analyse
than most bacteria. So it is not surprising that in contrast to Rhizobinrn, much less is
known of the role that plant genes play in the regulation of symbiotic nitrogen fixation.
Nodulins are specific proteins synthesised by plants during the morphogenesis and
functioning of the root nodules responsible for nitrogen fixation in the Legume-
Rhizobium symbiosis (Legocki and Verma, 1980; Van Kammen, 1984). It is clear
that a complex series of developmental steps, involving co-ordinated regulation of
plant and microbial genes, are required (see Nap and Bisseling, 1990) and to date,
nearly fifty mutations at loci involved with nodule development and function have been
identified using classical genetic studies. Unfortunately, no mutations have been
detected for any of the known nodulins.
69
It is only relatively recently that molecula¡ cha¡acterisation of genes involved with the
legume-Rhizobium symbiosis has been made possible by advances in recombinant
DNA technology (Rolfe and Gresshoff, 1988; Scheres et aI., 1990; Lobler and
Hirsch, 1993; Miao and Verma, 1993). Already, a gene-for-gene complementarity-like
system (Flor, l97l) has been reported in legume-Rhizobiutît spp. associations
involving induced mutants of pea and lucerne (see Caetano-Anollés and Gresshoff,
1992). These plants may well provide excellent s)¡tems for molecular analysis of plant
genes involved in this symbiosis.
The use of induced mutations as an adjunct to investigations of fundamental plant-
resea¡ch questions will continue to grow. The "biological revolution" (Kerr, 1985) of
the last 15 years has presented researchers with the tools to detect mutations at the
molecular level and to analyse DNA sequences which may be responsible for the
observed variations. The establishment of reliable and more general transformation and
regeneration systems should eventually allow the elucidation of functions of cloned-
genes via complementation of suitably characterised plant mutants. Generating and
identifying plants which form a range of atypical mycorrhizal associations from
genotypes that are normally mycorrhizal (and preferably non-leguminous) is an
important research priority. Our understanding of the molecular and genetic processes
underlying recognition and compatibility between plant species and VAM fungi is
probably dependant upon a successful mutagenesis program.
4.2 Barley Mutagenesis
Barley remains one of the chief experimental plants for mutagenesis research. It is
diploid, self-fertilising, has a small number (2n = 14) of large chromosomes and is
genetically well characterised. These features have allowed the identification and
subsequent characterisation of an aray of mutants at the whole plant or cellular level
(Nilan, 1987; Nilan et al., 1981). Induced mutations in barley have been used to
elucidate the pathway of flavanoid biosynthesis (see Jende-Strid, l99l), modify
powdery mildew (Erysiphe graminis) resistance (Torp and Jgrgensen, 1986; Robbelen
70
and Heun, 1990), and improve agronomic traits such as early maturation, reduced
lodging and increased yield (see Sigurbjömsson, 1975). The choice of barley as the
host plant for these studies was greatly influenced by these cha¡acteristics (also see
Chapter 3) and because of the considerable volume of work describing approaches to
barley mutagenesis. The choice of mutagen, also a very important consideration, is
discussed in detail in the next section.
7I4.3 Mutagenic Agents.
(Ð Chemical mutagens
The number of chemical mutagens is large but only a few are useful. The effective
ones ¿ue classified into four groups: base analogues, antibiotics, azides and alkylating
agents.
Base analogues, such as S-bromo-uracil (BU), 2-amino-purine, 8-ethoxy caffeine and
maleic hydrazide (MH), are chemically related to the DNA bases, adenine, guanine,
cytosine and thymine. They appear to be incorporated into DNA without hindering
replication (see Heslot, 1977), but because of thei¡ modified stuctures they are able to
induce pairing errors (base changes) and chromosome breaks. Both BU and MH have
been identified as effective plant mutagens (Jacobs, 1964; Darlington and Mcleish,
1951). A number of antibiotics such as actinomycin D, mitomycin C and streptoniogin
have also been found to possess mutagenic properties (see Heslot, 1977). The
usefulness of base analogues and antibiotics as mutagens is limited because the
frequency of mutation is so low as to make screening impractical.
Azides (N3-), particularly sodium azide (SA), can generate mutations at high
frequencies with negligible chromosome aberrations (Kleinhofs el al.; 1974). The
mode of action of sodium azide is still not well understood (Nilan, pers. comm.) as
azide itself is not mutagenic and is metabolically converted by O-acetyl-serine(thiol)-
lyase to the compound b-azidoalanine, as neither racemer of b-azidoalanine is
mutagenic, fufher biochemical activation must occur (Nilan, pers. comm.).
Alþtating agents such as ethyl methanesulphonate @MS), diethyl sulphate and methyl
nitroso urea form the largest and most important class of mutagenic chemicals. Like
SA, they are very effective plant mutagens, characterised by high frequencies of
induced mutation with minimal chromosomal aberrations and low frequencies of lethal
genotypes (Gaul et a1.,1972; Gottschalk and rùVolff, 1983; Nilan et a1.,1981). They
a¡e also easy to prepare and use but must be handled with caution. When these
72
compounds (and SA) are used in conjunction with certain modifying factors such as
presoaking in H2O or with certain metal ions (ie. Zn2+ or Cu2+) or dimethyl
sulfoxide (see Heslot, 1977), they are extremely efficient at inducing mutations. These
mutagens react with DNA by alþlating both the phosphate groups associated with the
deoxyribose and the pyrimidine and purine bases. Some recent data suggests that EMS
may be capable of inducing structural alterations in plant genes (deletions) in addition
to the point mutations usually ascribed to alkylating agents (Okagaki et aI.,l99l).
(ü) Physical mutagens
The most successful and widely used physical mutagens are gamma rays, X rays, fast
and thermal neutrons. These types of radiation can release discrete units of energy,
ionizations or ion pairs, as they pass through organic matter. These ionizing events can
cause chemical changes which are capable of inducing single and, more importantly,
double strand chromosome breaks. These may lead to structural re-arrangements (see
Salganik and Dianov, 1992) of chromosomes (ie. translocations, inversions,
duplications and deletions) and the subsequent loss or gross alteration of gene
expression and protein function.
X-rays and gamma-rays a¡e forms of electromagnetic radiation that have been widely
used as mutagens. X-ray machines accelerate electrons which are then stopped
abruptly by a target (usually tungsten or molybdenum) to produce a wide range of
radiation wavelengths of variable energies (50-300 keV). In contrast, the wavelengths
of gamma-rays are far more homogenous than X rays and, being of shorter
wavelength, they possess more energy/photon (up to 4 MeV). Monoenergetic gamma
radiation is also relatively easy to obtain from radioisotope sources such as 60Co and
1379s. While it is possible to filter out the longer wavelengths generated by-X ray
machines and leave only the penetrating high-energy radiation, this type of X-ray
source can be expensive and difficult to access (for a complete discussion see Briggs
and Constantin,1977). For practical reasons, gamma rays are often preferred to X-
rays.
73
Neutrons have been shown to be highly effective plant mutagens. Fast neutrons (0.5-2
MeV) are usually obtained ¡s¡1235[J reactors undergoing nuclea¡ fission. By using
moderators, such as carbon and hydrogen, the energy of fast neutrons can be reduced
to about 0.025 KeV to produce thermal neutrons. Initially, the lack of adequate and
uniform dosimetric techniques (Smith, 1961) hampered the use of neutrons in
mutagenesis studies. Increased public accountability of nuclear facilities has stimulated
nuclear resea¡ch in recent years and as a consequence, many of the early dosimetry
problems (again see Briggs and Constantin, 1977) have been solved. A major
difficulty with using this type of radiation for mutagenesis is gaining access to a
neutron source.
4.4 Choice of Mutagenic Agent
This mutagenesis study was part of a molecular genetic investigation of mycorrhizal
interactions. The aim was to generate stable barley lines that did not form mycorrhizas
or formed abnormal mycorrhizas. One of the long term aims is to use myc- plants in
physiological and molecular investigations in order to ascribe functions to
morphological structures. (eg. arbuscule-less mutants could be used to assess the
importance of the a¡buscule in nutrient transfer between the symbionts) and cloned
symbiosis-related genes. These aims were kept very much in mind when deciding
what type of mutagen was to be used. Practical issues were also a major consideration.
The vast majority of chemically induced mutations are thought to contain single DNA
base pair changes with commensurate amino acid substitutions (see Gottschalk and
Wolf, 1983). The translated protein is either inactive or has reduced and/or altered
biological activity. These genotypes provide excellent material for plant breeding and
biochemical studies. Unfortunately, they may not be particularly useful for molecula¡
analyses because very few point mutations will affect transcription of the mutated gene
or the transcript length of the mature mRNA. Gross effects a¡e almost essential for the
detection and isolation of specific genes using cunent recombinant DNA technology.
74
Physical mutagens also have practical limitations: It can be difFrcult to access suitable
sources of radiation and a high frequency of lethal mutations can be expected (R.
Lance, pers. comm.), necessitating a large initial Ml population. However, they often
induce the types of mutations detectable using recombinant DNA or cytological
techniques.
In the experiment described here gamma radiation was used as the mutagen because a
60Co source was available, courtesy of the Royal Adelaide Hospital, and the barley
breeders at the Waite Agricultural Reserach Institute provided the technical assistance
and field space to plant out a large population of gamma iradiated barley seed.
4.5 Generation of M2 Populations from 60Co
Gamma-ray Irradiated Hordeum vulgare cv. 'Galleon' Seed.
Two Ml populations (6000 seeds each) of gamma irradiated Hordeum vulgare cv.
"Galleon" seeds were prepared using a 60Co radiotherapy unit at the Royal Adelaide
Hospital which emmitted approximately 650 Rads of gamma-rays/minute. The seeds
were irradiated in l0 x l0 x 5cm clear perspex, open top containers which allowed an
even, shallow (lcm) spread of the seeds. Population I (Pl) received 2xL5 minute (20
kRads total) exposure to the gamma-rays and population 2 (P2) received 2 x 19 minute
(25 kRads total) exposure to the radiation. The initial and final doses were given 24
hours apart to ma;rimise the efficacy of the treatment and the seeds were thoroughly
mixed and carefully respread over the trays between the treatments to average out the
dose for each seed (Borojevic et aI., t977). The irradiations were carried out at room
temperature (22oC).
Approximately 3500 seeds from each population were planted out in separate field
plots in May 1990, using a direct-drilling seeder. These seed plots were separated
from the Waite Institute barley breeding lines by a2metre gap of bare ea¡th which was
in turn surrounded by a 2 metre wide apron of parental H. vulgare cv. "Galleon"
plants. The Ml plants developed normally through the growing season and received
75
no fertiliser or pesticide treatments. In February 1991, the plants were harvested
individually and the M2 seeds collected from each plant were placed in brown paper
bags and numbered.
Both populations had high moftality rates; approximately 25Vo of Pl and 80Vo of P2
failed to germinate or mature. A further 2.5Vo of P2 failed to set viable seeds. These
data indicated that the gamma irradiation treatments had induced mutations in both
populations but also that the large amount of genetic damage to P2 may have reduced
the usefulness of that population. The high mortality rate had only left 279P2 families
for screening, which may have been too few (statistically) to detect useful mutations.
The results obtained from the growth and colonisation experiments (see Chapter 3)
showed that preliminary screening for non-mycorrhizal 'Galleon' plants could not be
done using phosphate deficiency symptoms. This greatly reduced the rate at which the
plants could be screened because the only way of determining the mycorrhizal status of
the plants in the mutagenised populations was to check the roots of each plant
individually.
76
4.6 Screening of the Gamma-
irradiated Hordeum vulgare MÍl Populations
for myc Mutant Phenotypes
Twelve seeds were selected from each family for screeningl this number of seeds gives
a 95Vo probability of detecting homozygous recessive mutations. The twelve seeds
from each M2 family were sterilised, germinated and planted out in a small plastic pots
containing 250 grams of potting soil and lÙVo Glomus intraradices pot inoculum as
described (2.3). The seedlings were grou,n for 21 days in the greenhouse before each
plant was non-destructively sampled. Root samples were taken using a 1 cm corer,
and used to determine the extent of mycorrhiza development within roots (2.5).
Initially, arbuscular development was assessed subjectively. Putative a¡buscule
mutants were then reassessed quantitatively; the root pieces were re-scored for the
presence of G. intraradices and the presence of arbuscules within the segments.
Arbuscular development was expressed as a proportion of total colonisation. The
extent of colonisation and a¡buscular development within the M2 roots was then
compared with that of roots collected from four parental 'Galleon' plants grown at the
same time and under the same conditions. Plants which varied from the parental
'Galleon' controls by more than25Vo for total root colonisation or had not developed
arbuscules were repotted into a rich soil mix and grown on to collect the M3 seeds.
In all, forty M2 families from Pl and one hundred and twenty M2 families from P2
(for a total of 1880 plants) were screened in this manner. Four of the P2 families (FI1,
Fl4,F52 and F144) continued to segregate for lethal chlorophyll (albino) mutations.
A total of 34 putative rnyc- mutants from sixteen families were identified in the initial
screening (see Table 4.1). All of the putative mutants were from the P2 population.
Six families (F4, F24,F25, F6l, F62 and F104) contained three or more plants which
appeared to form abnormal mycorrhizas. In the majority of these putative zryc plants,
the extent of colonisation was lower than the parental 'Galleon'. Three plants from
77
F24 and one from Fl09 had very little a¡buscule development and two plants from
F13l were exEemely heavily infected in comparison to the conEol plants.
Table 4.1
List of putative myc barley mutants identified from primary screening of 1880 Mlplants. Twelve seedlings from each of 160 Ml families of Hordeum vulgare cv.'Galleon' (obtained by 20 or 25kRad gamma-irradiation) were inoculated with the VAmycorrhizal fungus Glomus intraradices . The roots were non-destructively sampled
after 2l days growth and assessed for mycorrhizal colonisation (Vo of the root length
sampled). The plants which va¡ied in extent of colonisation from parental controls by
more than 25Vo or appeared not to have formed arbuscules (shown with *) a¡e listed.
(vo) CONTROL PLANTSVo
F4 444-8
4-104-121t-ltl4-824424-524-tt25-l25-5
25-1025-t252-56t-36t-76r-962-362462-9
62-1062-tt67-791-29r-597-9t0l-6t@'-3l0/,-9
104-10109-613t-7t3t-12t4ø.-10
872999
252024t2ll9t2l9t21lt481l691113L4t41515llT3t22t55487
-46-53-86-40-43-43+25
0+2O-40-45-55-40-25-33-38- ')')-55-38-66-50-38-27-26-26-21-32-50-41-45-5
+ 175+ 140-65
FllFl4
F24 *
F25
F52F6l
F62
F67F9t
F97F101F104
Fl09 *Fl31
Ft4ø.
78
The M3 seeds from each of the putative myc- mtJ¡arnt lines (six per line) were
rescreened as described above except that the whole root systems were harvested and
weighed. The fresh weights were used as an indicator of variances in total root length
for each M3 line. The mean fresh root weight of one M3 family (F4), was 2lvo lower
than the 'Galleon' control plants. The root systems of three plants from this M3 line
appeared to be much smaller than the others and were also morphologically different
from the control plant root systems; they were stunted and slightly thickened. The rest
of the putative myc- farnäes did not signifrcantly differ in mean fresh root weight from
the control plants.
The extent of mycorrhizal colonisation was not significantly different, quantitatively or
qualitatively, in the great majority of putative mutants when compared with the parental
'Galleon' in the second screening. However, one line (F25-10) showed a small but
significant decrease in the extent of mycorrhizal colonisation within its root system
(see Figure 4.1). The two putative myc mutants, F4-10 and F25-10 were repotted into
a rich soil medium and grown on in the glasshouse. The seeds (M4) were collected
and stored.
Line F4-10 segregated into two distinct sub-populations, one (F4-l0A) contained 3
plants which were very poorly infected. These plants had much lower root fresh
weights (35Vo compared to the controls) and displayed the unusual root morphology
described above. The mean colonisation percentage and the root fresh weights of the
other sub-population (F4-l0A) was not significantly different from the parental
controls. A graph showing the extent of colonisation for F4-104, F4-10B and F25-10
compared to the parental 'Galleon' conEol plants is shown in Figure 4.1.
The M3 line, F25-10, appeared to be phenotypically the same as parental 'Galleon'. It
grew notmally and set seed. The F4-l0A type plants began to display unusual shoot
morphology and leaf chlorosis indicative of nutrient deficiency after 14-17 days
growth (see Figure 4.2). All such plants became necrotic and died without setting
seed. A further trvo F4-104 type plants were identified from ten F4-10 seeds grown as
79
described above. These plants were passed on to the Dept. of Plant Science glasshouse
manageress in the hope that she could keep them alive. They showed the same unusual
shoot morphology and nutrient deficiency symptoms and died within 30 days of
germination.
Both of the putative nyc mutants formed myconhizas but at significantly lower levels
of colonisation under the growth conditions described above. Both phenotypes
contained all the structures associated with normally developed and functional G.
intraradices - H. vulgare mycorrhizas. It appeared that the putative mutations did not
involve mechanisms mediating compatibility between the symbionts (ie. resistance
reactions) or the formation of the gross stnrctures. No ultrastructural analyses were
carried out and alterations to the fungal - plant cell wall interfaces cannot be ruled out.
However, it seems unlikely that mutations to the ultrastucture of mycorrhizas could
affect overall development to such a large extent. Presumably the plant genes involved
in these stages of mycorrhiza formation are determinants of functional (eg. nutrient
transport) mechanisms rather than compatability and gross fungal morphology.
The gene mapping projects in the laboratory of Dr. P. Langridge, which were being
caried out concurrently with this research, had indicated that some seed stocks of the
barley cultivar 'Galleon' \ilere not pure. Polymorphisms were detected between and
within plants expected to be homozygous. It is possible that early stocks of 'Galleon'
seed may have been contaminated with another cultiva¡ or that pollen drift from other
cultivars had resulted in cross breeding rather than selfing.
The comparison between the barley cultiva¡s 'Schooner' and 'Galleon' (see Chapter 3:
Experiment l) showed significant differences between the two genotyptes with respect
to their response to and extent of mycorrhizal colonisation. As discussed previously,
this is in agreement with reports which describe similar differences between cultivars
of many other species. It is possible that the differences in the extent of mycorrhizal
colonisation observed between parental 'Galleon' and the M2 lines 4-104 and
80
particularly 25-lO, are a result of genotypic differences caused by 'genetic
contamination' and not by induced mutation.
FIGURE 4.1
Extent of colonisation by the VA mycorrhizal fungus, Glomw intraradices, of
Hordeutn vulgare cv. 'Galleon' and three putative mutants derived from gamttt¿-
irradiated 'Galleon' (F4-104, F4-10B and F25-10). The values represent the means
of six samples taken frrom the roots of six plans of 'Galleon' and the putative mutant
F25-10 and three samples taken from theroos of three plants of the putative mutants
F4-104 and F4-108. Error bars represent the standa¡d errors of these means.
Figure 4.1,
Éo^duØ OL)
EP€¡r
o(d!P(HV
H
.g BeX.vrq
25
20
5
0
30
5
0
1
1
GALLEON F25-10 F4-1OB F4-1OA
Barley Line
FIGURE 4.2
Growth responses of the parent Hordeum vulgare cultivar 'Galleon' and rwo
putative mutants derived from gamma-irradiated 'Galleon' (F4-104 and F25-10) to
additional P (0.5 mmoles/kg) or inoculation with the VA mycorrhizal fungus Glomus
introrodices . The plants \rrere photographed afær 21 days growth.
(ì/il,l,ltoN-ill +l'
(ìr\l,l,EON
+l¡ -l''r-¡0
+Nl -l'{.t0
+tr1 -l)2.s- t0+i\f -P
25-10+i\l -t¡
NtYCìOtU{lIrZl\ t)tì¡'tclt.:N.t' (nfl ,c_)
ìu4 (;^t,t,l.:oN' t,tN¡ts
83
4.7 Polymerase Chain Reaction
(PCR) Analysis of the Putative Barley
myc- Mutants, 4-104 and 25-10.
Conventional varietal indentification is based on morphological description
(Stegemann, 1984). This is time consuming and not always accurate due to
environmental effects. Biochemical ma¡kers (ie. isozymes) are available which provide
quicker and more accurate assessments (see Cooke, 1988). DNA-based techniques
such as restriction fragment length polymorphism (RFLP) and PCR have also been
successfully used to detect polymorphisms in cereals (tWeining and Langridge, 1991).
The PCR-based techniques have two major advantages; they allow the detection of a
larger number of polymorphisms which helps to distiguish between closely related
va¡ieties; and they are cheaper, quicker and easier than RFLP analyses'
The polymerase chain reaction (PCR) involves the in vitro amplification of DNA
fragments using DNA polymerase (Saiki et al., 1985; Mullis and Faloona, 1987).
Short oligonucleotides (primers) complementary to the DNA flanking the target
sequence, hybridise to opposite strands of the DNA molecule with their 3-OH ends
pointing towards each other. The primers are enzymatically extended to synthesise
DNA complementary to each strand of the target sequence. Repeated cycles of heat
denaturation of the DNA, primer annealing and primer extension, result in the
amplification of the DNA sequence defined by the 5-P ends of the primers. The
extension products also serve as templates and essentially double the number of target
DNA fragments produced in each cycle. This results in rapid and exponential
amplification (up to several millionfold) of specific ta¡get sequences.
Several PCR-based approaches are available for consideration. The simplest methods
use the presence or absence of a DNA fragment band or band sequence length
polymorphisms as determinants. For example, it is possible to design (or purchase)
short oligonucleotide primers complementary to highly conserved DNA regions which
allow the amplifrcation of variable sequences which lie between them (Skolnick and
84
Wallace, 1938). Degenerate primers, coupled with lower annealing temperatures (see
Arnheim and Erlich,1992) or microsatellite primers (Litt and Luty, 1989; Beckman
and Soller, 1990) can also be used to produce DNA product banding patterns which
allow the detection of polymorphisms at the plant species and cultivar level (Morgante
and Olivieri, 1993; Rafalski and Tingey, 1993) .
Recently, Weining and Langridge (1991) described the use of primers, designated Rl,
E2 and H,to distinguish between a large number of barley cutivars and local (South
Australian) breeding lines. These three primers and four microsatellite primers (see
Appendix 6 for the sequences), which also produce polymorphic banding patterns for
barley cultivars usually grown at the V/aite Institute (4. Karakousis pers. comm.),
were used to generate DNA fragment banding patterns using template DNA extracted
fro*f ,
(Ð the putative myc- mutant lines 4-l0A and 25-10,
(ü) l0 plants grown from randomly selected seed of the
parental'Galleon' PoPulation,
(üi) a plant positively identifred as 'Galleon',
(iv) and the wheat variety 'Chinese Spring'.
The PCR amplifications were carried out as described (2.7.4.2) and the results are
shown in Figures 4.3 and 4.4. Figure 4.5 is an example of the polymorphic banding
profiles of amplified products produced by the primers H and AT when used with
genomic DNA extracted from several local barley cultivars. The banding profile was
different for each primer set and no polymorphisms were detectable between any of the
barley plants. 'Chinese Spring' was clearly distinguishable from barley for all primer
combinationr. T[*t t(r lir.os .^r¿ ^tl o.t (e'ael' p*]"triæl¡ 'G^ll.o^t'
As no evidence of cross pollination was detected using the PCR-based identification
method (cross pollinated plants, in effect heterozygotic Fl plants, should be clearly
discernable) it was concluded that the phenotypic differences observed between F4-10
85
and F25-10 were not caused by 'genetic contamination' of the original parental
'Galleon' seed. It seenrs most likely that the observed phenotypic differences can be
ascribed to the gamma irradiation treatment.
FIGURE 4.3
Agarose gel electrophoresis of polymerase chain reaction amplif,rcation products of
DNA from four barley lines and one wheat line using the primers AT, ST, 84, Rl,
LA and CP.
Lane 1 in each set of reactions is the parent Hordeum vulgare cultivar 'Galleon';
lane 2 is a plant positively identified as 'Galleon'; lanes 3 and 4 are the putative myc
mutants F4-104, and F25-10 respectively, which were derived by gamma irradiation
treatment of the parent cultivar; and lane 5 is the Triticum aestivum cultivar
'Chinese Spring'.
The DNA markers (shown as lane M in the first set of reactions) are Hind Illdigested ÀDNA. The 580bp marker bands are indicated at the right hand side of the
figure.
Figure 4.3
ATm12345
RI
ST
LA
E4
CP
- 580bp
- 580bp
-ttIo
ID
aO
l lIs
5
t
TilF=--CO
rlrtaI
FIGURE 4.4
Agarose gel electrophoresis of polymerase chain reaction amplification products of
four barley lines and one wheat line using the primers B 1, B I+F.4, E1 and E2.
Lane 1 in each set of reactions is the parent Hordeum vulgare cultivar 'Galleon';
lane 2 is a plant positively identified as 'Galleon'; lanes 3 and 4 a¡e the putative myc
mutants F4-104, and F25-10 respectively, which were derived by gamma irradiation
treatment of the parent cultivar; and lane 5 is the Triticum aestivum cultivar
'Chinese Spring'.
The DNA markers (shown as lane M in the first set of reactions) are Sau3{digested
pTZlSU. The 603bp marker bands are indicated at the right hand side of the figure.
Figure 4.4
B1m12345
E1
B1 + E4
E2
- 60jbp
- 603bp
ll= E n:-r.¡ ¡
=¡L lr
¡!,---D
--Il----
O
H-ata--
IO
FIGURE 4.5
Agarose gel electrophoresis of polymerase chain reaction arnplifrcation prcducts of
five barley lines and one wheat line using the primers AT, ST, E4, Rl.
Lane 1-5 in each set of reactions a¡e the Horfuutn vulgare cultivars 'Schooner',
'Clipper', 'Galleon', 'Haruna Nijo' and 'Betzes' repectively, and lane 6 is the
Triticwn aestivutn cultiva¡ 'Chinese Spring'. Polymorphic bands ate indicated by
the arrow heads.
The DNA markers (shown as lane M in the first set of reactions) are Hind Lttdigested
^.DNA. The 580bp marker bands are indicated at the right hand side of the
figure.
Figure 4.5
ATm123456
E4
ST
R1
- 580bp-t--
I r
¡t
t
53
sa). .--Ê
:E==s----
!FE---F ----
!!
89
4.8 Discussion and Conclusions
The high mortality rate observed for the Ml seed of P2, suggests that considerable
genetic damage occurred to the seeds which received the25 kRad gamma irradiation
treatment; 807o of the mutations were lethal. The Ml seeds of Pl were less affected (in
terms of lethal mutations) than those of P2. No direct comparisons can be made to the
parental population survival rate because 'Galleon' mortality rate was not measured at
the time. However, data from subsequent experiments indicated a survival rate of over
95Vo.
The PCR analyses of the original 'Galleon' population and the 2 puøtive myc- mutants
did not indicate any 'contamination'from another genotype, so it seems probable that
the phenotypic variations observed between 'Galleon',4-104 and 25-10 with respect
to the extent of mycorrhizal colonisation, arose from induced mutation(s).
The lower colonisation percentage in F25-10 may reflect slower development within
the root (caused by slight differences in nutrient availability and/or transfer) or a lower
number of initial colonisation points (indicating greater resistance to root penetration).
Line F25-10 was only slightly less susceptible to VA mycorrhizas than parental
'Galleon' and because there were no obvious differences in the pattern of
development, experimental error cannot be ruled out as the source of the observed
va¡iation. A larger sample of this M3 population needs to be critically evaluated with
regard to the extent and development of colonisation.
The F4-l0A line segregates for a lethal mutation in the M3 population (11:5). The
population tested is too small to accurately determine the segregation ratio; it may be
2:l or 3:1. The low levels of normal colonisation in F4-l0A established that VAM
colonisation is possible in this plant line and that in all likelihood, the mutation
responsible for this phenotype does not involve genes controlling structural changes to
plant cells during the establishment of the symbiosis. The chlorosis and subsequent
necrosis of leaves in F4-104, after what appears to be normal early growth, may be
90
indicative of the slow 'starvation' of this genotype following the exhaustion of seed
nutrient reserves. This phenotype could be the result of radiation induced changes in a
gene encoding a component of the epidermal root-cell wall. Neumarln et al., (1994)
have reported root cell hardening and root inhibition in response to salinity stress.
They postulated that biochemically regulated changes to the visco-elastic properties of
cell walls (see Passioura and Fry, 1992) may be responsible for the limited uptake and
supply of essential nutrients associated with salinity stresses (Zidan et a1.,1992). An
induced mutation affecting cell wall visco-elasticity could render these cells not only
impermeable to soil solutes but almost totally impenetrableby Glomus intraradices.
Arabidopsis thaliana has become the major experimental plant for molecula¡ and
genetic analysis of plant development and function. It is amenable to large scale
screening and has a small genome, facilitating ch¡omosome walking (see Koornneef,
l99l). Unfortunately, it is one of the few non-mycorrhizal plants. Myc- mt'úatrons had
previously been identifred in pea (Duc et a1.,1989) but they appear to be linked with
nodulation (nod- ) mutations which may have complicated the results, not to mention
the difficulties in setting up a model system involving two micro-organisms. Barley
was used for this work because it appeared to be a good compromise for molecular,
genetic and mutant screening.
Yields for single gene mutations in A. thaliana can be in the order of 1 mutant per
40,000 (Cabrera y Poch et al., 1993) and even lower for barley (see Briggs and
Constantin,lgTT). The number of families screened was not large (150) but the yield
of myc mutants (1.3 mutants per 100 families) suggests that a large number of genes
may affect the rate and./or extent of root colonisation in barley by VA mycorrhizal
fungi. Isolating a large number of mutants defective in mycorrhizal development may
not be particularly difficult. The two putative myc- mltant lines, F25-10 and F4-104,
are less susceptible to VA mycorrhizal colonisation than parental control plants but
neither appear to be developmental or functional mutants. Identifying the genes which
directly control the interaction will prove much more difftcult.
9I
Chaoter 5
5.1
Molecular Cloning and Isolation ofBarley -mycorrhiza Related Plant cDNA's
Introduction
Elucidating the molecular mechanisms underlying the VA mycorrhizal symbiosis
will require both biochemical and molecular genetic data. Making direct biochemical
comparisons between mychorrizal and nonmycorrhizal roots is difficult because
individual VA mycorrhizas are almost impossible to detect and excise without
staining, and so whole root systems (which contain uncolonised sections of root
tissue and VA mycorrhizas of varying physiological stages) must often be used in
order to obtain enough material. This may confound the results obtained, as do the
physiological and biochemical changes induced by the improved P uptake usually
associated with mycorrhizal roots (see Chapter l).
These limitations also apply to current molecular genetic methods but to a lesser
extent. Techniques for investigating temporal and spatial gene expression within
plant tissues are extremely sensitive relative to most biochemical assays; this means
that smaller amounts of root tissue can be used for most nucleic acid analyses.
Broadly speaking, the molecular genetic stategies aimed at isolating genes can be
classified into two groups: Targeted and Non-targeted.
92
5.1.1 The Targeted Approach to Gene Cloning
The targeted approach requires the knowledge (or at least the assumption) that
specific genes are involved. Suitable genomic or cDNA libraries can then be directly
and rapidly screened with protein or heterologous DNA probes obtained from other
sources to isolate the corresponding gene(s). The VA mycorrhizal symbiosis is still
poorly understood at the molecular level, nonetheless the biochemical and
cytochemical evidence (again see Chapter 1) suggests there are potential candidate
genes for direct screening; these could include genes which encode proteins involved
in cell wall and cytoskeleton synthesis (DexheimeÍ et al., 1979; Scannerini and
Bonfante-Fasolo, 1979; Bonfante-Fasolo and Grippiolo, 1982; Bonfante-Fasolo ¿l
al., l99O) and sugar metabolism (see Smith and Gianinazzi-Pearson 1988), and
genes which encode ATPases (Marx et aI., 1982; Gianinazzi-Pearson et al., 1991).
The major drawback of this approach is that it is unlikely to reveal new or novel
features of the system under analysis: It can only confirrn or refute previous
thinking.
5.1.2 The Non-targeted Approach to Gene Cloning
The non-targeted approach does not require any knowledge of gene function or
involvement. It relies on the increasingly sophisticated techniques available for
modifying, manipulating and identifying nucleic acids in vitro and invivo. There are
several potential strategies:
(Ð'Shotgun' CloningandTransformation.
This approach involves the cloning of complete and functional genes from partially
digested genomic DNA and the subsequent transformation of specific (typically
mutant) phenotypes with those clones. Monitoring the transformants for altered (or
restored wild-type) phenotypes can allow elucidation of the biochemical and/or
93
physiological functions of the transforming genes. 'Shotgun' cloning stategies have
proven successful in isolating genetic sequences and confirming gene functions in
several bacterial and fungal species (for examples see Nester et al., 1984; Djordjovic
et a1.,1987; Staskawicz et aI., t984; Leong and Holden, 1988).
Cloning plant genes by complementation in suitably cha¡acterised plant phenotypes
is logistically difficult if not impossible. Plant genomes a¡e orders of magnituide
larger than those of bacteria and fungi, necessitating very large libraries of clones
(see Brown, 1991) and the efficiency of transformation in even the most amenable
plant species is extremely low. Complementation of plant DNA sequences in
suitable E. coli and S. cerevisiae phenotypes has been used to identify several plant
genes (see Gibson and Somerville, 1993). Whilst this approach appears suitable for
cloning genes involved in basic cell metabolism, it is unlikely to prove useful for
identifying genes involved in plant developmental processes (ie. cell differentiation)
or plant-microbe interactions.
(ii) Gene tagging
Many genes a¡e dominant and the insertion of DNA sequences into them should lead
to a loss of gene function. Gene tagging involves the creation of a population of
mutant plants by the random insertions of a transposable elements (see Balcells er
al., t99L) or Agrobacterium -mediated T-DNA (see Koncz et a1.,1989; Feldmann,
1991). The target-gene sequences of interest which flank the transposable element or
T-DNA sequence can then be isolated by using the inserting sequences as a
hybridisation probe. Insertion-sequences have been used to identify and isolate,
amongst others, nitrogen fixation and nodulation genes from Rhizobium species (see
Djordjevic et al., 1987), genes of the anthrocyanin pathway from maize (Federoff er
a1.,1984), and genes which affect trichome (GIABROUSI ) and chloroplast (ch-42 )
development in Arabidopsis thaliana (Marks and Feldman, 1989; Koncz et aI.,
1990). The insertional mutation approach has several problems which will only be
94
very briefly discussed here (for more complete discussions see Ellis et aI., 1988;
V/alden et al., I99l;'Walbot, 1992).
Only two plant species, Zea rnays L. (maize) and Antinhinum majus L., have
genetically well defined and cha¡acterised transposable elements. To overcome this
obvious limitation, transgenic plants containing an active transposable element from
muze have now been obtained in several species including A. thaliana (Van Sluys
et aI., 1987), carrot (Van Sluys et a1.,1987), tobacco (Baker et al., 1986), tomato
(Yoder, 1990), and rice (Murai et aI., l99l). Clearly progress is being made but
many important crop species (including barley) are not yet amenable to this type of
genetic manipulation.
lMallze is also subject to very high natural mutation frequencies. As Ellis et al. (1988)
point out, the number of background mutations detected for each insertional
mutation could range from 33 to 4000 in a population considered large enough to be
"reasonably sure" of obtaining at least one transposable element-derived mutation
event. Maize also contains multiple copies of sequences which hybridise to the
transposable elements suitable for this type of experimental approach. Separating the
background mutations from the insertion events and characterising the many false
positives that are likely to be detected when transposable element sequences are used
as probes is impractical for any but large teams of researchers.
Agrobacterium- mediated transformation (T-DNA insertion) shows considerable
promise as a method for gene-tagging in dicot plant species which are most
amenable to transformation and whole plant regeneration techniques (A..thaliana,
tobacco and tomato). Transformation and plant regeneration is not yet efficient
enough in monocot species (with the possible exception of rice; see Christou et al.,
1992) for this type of approach to be given serious consideration for this project.
95
(iii) Differential screening.
Complementary DNA's (cDNA's) are doublestranded DNA copies of the messenger
RNA (mRNA) molecules found in cells which serve as templates for protein
biosynthesis. Apart from non-coding regions at the 5'and 3'ends, eukaryotic mRNA
is free of the non-coding regions (introns) which usually disrupt the coding regions
(exons) of eukaryotic genes (Darnell, 1982).
Thus cDNA is a source of intron-free genetic information. This is a necessity when
the aim of gene cloning is to produce eukaryotic-protein expression in a bacterial-
cell system. Bacterial genes do not possess the characteristic intron-exon structure of
eukaryotic genes and appear not to have the mechanisms for post-transcriptional
processing of RNA's containing internal non-coding regions. Even if protein
production is not an express aim, cDNA's are still an extremely useful source of
genetic material; cDNA clones are usually much smaller than their corresponding
genomic clones making cloning, manipulation and subsequent analyses of the DNA
sequences easier and more efficient; they represent only the expressed genes and this
is only a very small percentage of the total genomic DNA; and they make very
useful probes for hybridisation experiments, particularly in gene-expression studies.
Differential screening essentially involves comparing the relative abundances of
individual mRNA species in different RNA populations. A cDNA library is made
from one population of RNA, the individual clones are plated out, transferred to a
membrane in duplicate and hybridised with single-stranded labelled cDNA obtained
by reverse transcription of RNA from the original plant material and the material
being used for the comparitive analysis. By comparing the relative signal intensities
of clones within and between the two populations it is possible to draw conclusions
regarding the levels of expression of an individual clone in each of the two
populations. This approach has allowed the cloning of many tissue specific,
96
developmentally regulated and symbiosis-associated plant genes from a range of
plant species (see Table 5.1 below).
Table 5.1
Genels)
Auxin-induciblegenes.
Calmodulin-like gene
Cotton fibredevelopmental genes
Light-induciblegene.
Grain-developmentgenes.
Gibberelic acid-induced endochitinase
Pathogen-induced genes.
Myo-inositolsynthase Citrusleaves
Tobacco leaves
Cotton fibres
Rice shoots.
Aleurone layer ofbarley seeds.
Aleurone layer ofbarley seeds.
Erysiphe graminis -infected barley leaves
Nematode infectedpotato roots
Maize pollen.
Monke and Sonnewald, 1995.
John and Crow, 1992.
de Pater et a1.,1990.
Olsen et aI., L990
Swegle et a1.,1989.
Abu-Abied and Holland, 1994.
Davidson et a1.,1987.
Examples of plant genes isolated by differential screening
Plant Species and Tissue Reference
Strawberry receptacles. Reddy et a1.,1990.
Gurr ¿r al.,I99l.
Pollen-abundantputative actin-depolymerising gene
Rozycka et al., 1995.
Nodulin genes. Pea root nodules. Govers et a1.,1987.
Fruit-developmentgenes.
Ripe tomato fruit. Pear et a1.,1989.
The differential screening approach was selected for the analysis of mycorrhiza-
related genes for several reasons; this method has been succesful in isolating
symbiosis:pathogenesis-related plants genes from both pathogenic and mutualistic
systems; it is not derivative of other research and so there is the potential to isolate
97
and identify novel genes; and the methods required are technically well developed
and readily adaptable to the Glomus intraradices -barley model system.
5.2 Construction, Characterisation and
Differential Screening of a cDNA Library Prepared from
Mycorrhizal Hordeum vulgare cv.'Galleon' Roots.
The aims of this series of nucleic acid manipulations were:
(1) to constuct a cDNA library from 12-14 day old mycorrhizal
(M+) barley root tissue;
(2) to differentially screen this library with q-32P-dCTP-labelled
(radio-labelled) cDNA reverse transcribed from M* and non-
mycorrhizal (NlvI) 14 day old barley roots;
(3) to identify plant cDNA clones which detect mRNA's that
increase or decrease in abundance during the establishment of
the symbiosis and;
(3) to confirm the results of the differential screening using
DNA: RNA (Northern) hybridisation techniques.
5.2.1 Synthesis of cDNA
Fofy H. vulgare cv. 'Galleon' seeds were sterilised and germinated as described
(see 2.2). Twenty four germinated seeds of approximately equal size and
development were selected; half were transplanted into pots containing 400 g of
sand:soil (1:9 w/w) growth medium and the remainder were transplanted into pots
containing 400 g of growth medium inoculated with Glomus intraradices pot
culture (líVo w/w). Nutrient solutions A-E were added (10 rnl/kg of growth
medium) to each pot and the plants were watered to weight (l2%o wlw) every 2 days
(see Appendix l).
98
After 14 days growth the whole root systems of 3 plants from each treatment were
ha¡vested and assessed for G. intraradices infection (2.3): None of the NM plants
were infected; the extent of the infections in M+ plants were 16%o, lTVo and 2tVo.
The roots of all remaining plants and the leaves of the M+ plants were harvested,
carefully washed free of the growth medium and snap frozen in liquid N2.
The results described in Chapter 3 show that the majority of VAM development is in
the first few lateral roots of 14 day old'Galleon'plants. Accordingly, the bottom half
of each root system was excised and discarded. The root samples of each treatment
were then pooled, weighed and divided into 3 approximately equal portions (100-
150mg). Total RNA was extracted from each of the 6 individual subsamples as
previously described (2.7.2.L). The yield of RNA from each subsample was
determined by spectrophotometry (Maniatis et al., 1982) and lpg of each sample
was separated on an ribonuclease (RNAse) free I.57o TAE agarose gel, stained and
photographed (2.7.2.4). Non-denaturing electrophoresis was used because of the
improved staining of non-denatured RNA (which is usually associated with
secondary structures) in comparison to denatured RNA. The electrophoresis gel (see
Figure 5.1) indicated only slight degradation of the ribosomal RNA in all samples;
this can prossibly be attributed to low levels of exogenous RNAse activity in the
electrophoresis tank and gel and/or residual endogenous RNAse activity in the
samples which is not inhibited by the buffering conditions (RNA samples run in less
shingentþ prepared RNAse free systems were almost invariably more degraded than
those run in very strictly prepared RNAse free sytems).
The yields from the total RNA extractions (see Figure 5.2) showed that the
concentration of RNA (pglmg f.w.) was higher in M+ root tissue than in NM root
tissue. This.observation was supported by a considerable amount of data collected
over the course of this project. The fresh weights, extents of infection, and RNA
yields were determined for all tissues used in paired extractions (NM and M+) of
99
approximately the same fresh weight: Total RNA yields were always greater from
M* roots than NM roots when the extent of total infection was higher than lOVo.
This could indicate increased gene expression in the M+ root tissues due to in vivo
transcriptional activation and protein translation which corroborates the data reported
by Pacovsþ (1989) and V/yss et aI., (1990) showing that protein levels are higher in
Mr roots than in NM roots. However, some recent unpublished results suggest that P
starvation may increase RNase activity in plant roots (communicated by S. Smith)
which could lower totat RNA concentrations in NM roots by increased rates of RNA
degradation. The observed increase in RNA yield could also reflect higher
concentrations of RNA in the fungal hyphae relative to the barley roots. As the
proportion of fungal tissue is unlikely to exceed l07o of the total fresh weight in
even the most heavily colonised mycorrhizal roots (see Harley and Smith, 1983),
the RNA concentrations within the colonising hyphae would need to be at least 24
times higher than that of the root tissues to account for yield increases ranging from
lOA}Vo. A combination of all three explantions cannot be ruled out.
5.2.2 Construction and Characterisation of a M+ cDNA Library
RNA sample 6 was selected as the material for the synthesis of cDNA (see Figure
5.1). A cDNA cloning kit was obtained from Pharmacia; this kit utilises Moloney
Murine Leukaemia Virus (M-MLV) reverse transcriptase, which unlike Avian
Myeloblastosis virus reverse transcriptase, is not inhibited by ribosomal RNA
(Kotewicz et a1.,1935). Consequently, total RNA preparations rather than purified
poly A+ RNA can be used as the template for lst strand synthesis. This mitrhheduce
the total number of clones obtained but may increase the number of full length
cDNA's as the poly A+ RNA purification step involves buffering and temperature
conditions which may subject the RNA to further degradation by any contaminating
RNAses.
100
The cDNA synthesis was ca¡ried out using 50¡rg of RNA template as described by
the manufacturer except that CI,-32-P-dCTP (3pC) was added to the first strand
reaction mix. Samples were taken from the reaction mix after completion of the first
and second strand reactions (5pl and 10¡tl, respectively) and separated on an RNAse-
fuee 1.57o TAE agarose gel which was then dried down in a vacuum oven and
autoradiographed for 4 hours at -80"C. The developed autoradiograph (see Figure
5.3) indicated that cDNA:RNA hybrid molecules and doublestranded cDNA's
ranging from 200 bp to 4000 bp had been synthesised.
The cDNA's were ligated with Eco Rl/Nor I linkers, cloned into the EcoRL site of
the insefion vector l,$ l0 (Huynh et a1.,1984) and packaged into infective phage
using extracts supplied by Promega. Two inserlvector ratios were used for the
ligations (the molar ratios were unknown because the yield of cDNA was not
known) and l,gt 10 arms (lpg) were ligated with 0.5pg of Eco Rl digested pSP64
plasmid (supplied by Promega) as a ligation control. All ligations were ca¡ried out as
previously described (2.7.3.10). Undigested non-recombinant ì,gt10 DNA (l00ng)
was used as a control to test the efficiency of the packaging reaction. The 2 cDNA
ligations and the pSP64 control ligation were used to transfect E. coli host strain
C600HflI which carries a high frequency lysogeny mutation (IIl 4150).
Recombinant Àgt10 phage (which contains a disrupted hnm 434 gene) produce
plaques very efficiently in this stain whilst non-recombinants (with a¡ intact imm
434 gene) do not (see Winnacker, 1987). The non-recombinant lgtlO was used to
transfect E. coli host strain C600. The titres of the packaging reactions were
determined (see Table 5.2) as previously described (see 2.7.2.2) which indicated the
ligation and packaging reaction 1 contained approximately 6.6 x lOa cDNA clones.
The entire library was then plated out, amplified (Maniatis et a1.,1982) and stored as
lml aliquots.
lHliil t Åt i f:¡.Í $ t- I i1¡ì.,+,,,-p/'i'l
i i: l.i liïi: í:1Ì; t 1'í üi ¡ii-1,':il:r i [] Ì
Dilution 103 104 10s
Packaging control
Ligation control
Ligation I
Lieation 2
confluent
628
66
55
568
67
9
5
0
1
9
8
Table 5.2 Titration of cDNA library and control reactions.
Half of each packaging reaction mixture (l00pl final volume) was serially diluted
and 100p1 from each of the dilutions was used to transfect 100p1 of competant E. coli
which were then plated onto 90mm petri dishes containing LB medium. The plaques
were allowed to develop for 12 hours at37"C and then counted to determine the titre.
The estimated titre of phage in ligation l: = 66pfu x 103 - 6.6 x 105 pfu/rnl0.lml
= 6.6 x 104 clones in 100p1total volume
The quality of the cDNA library was checked by analysing the size of the cDNA
inserts using the Polymerase Chain Reaction (PCR) and hybridising a proportion of
the library with a ribosomal RNA clone to calculate the number of cloned ribosomal
sequences.
Two 12 base oligomer primers which anneal to the regions flanking the EcoRl site
into which the cDNA's were cloned (see Figure 5.4) were synthesised by D. Hein.
The insert sizes of fifty phage plaques were determined. The PCR's were carried out
as described(2.7.4.1) and the results (some of which are presented at Figure 5.5)
indicated that 957o of the clones contained inserts and that they ranged in size from
I2Obp to 3.5kbp. The majority of inserts were 450bp to 700bp in size.
r02
Approximately 10,000 clones were plated out (2,500 pfu/plate), transferred to
Hybond N+ hybridisation membranes and hybridised with a wheat rinbosomal DNA
clone (2-2.18). Less than lVo of the clones (56 clones out of approximately 10,000)
hybridised with the wheat ribosomal DNA (data not shown).
The PCR amplifications and hybridisation analysis indicated that the majority of the
cDNA's contained inserts of over 400bp in length and that less than 5Vo contained no
insert or ribosomal DNA. However, the number of clones obtained from the first
library was lower than expected, as good quality total RNA was readily available, a
second cDNA library was prepared and characterised as described above using
polyA+ RNA purified from 100pg of the M+ total RNA. The purification was
carried out using a Promega PolyATtract mRNA isolation kit. The second library
contained many more clones than the first (5 - 6 x t05) but the average insert size
was lower (200bp to 350bp; results not shown) and 45Vo of the phage contained
undetectable or no inserts.
FIGURE 5.I
Non-denaturing gel electrophoresis of total RNA (1pl from each sample/lane)
extracted, using a standardised proceedure, from 14 day old non-mycorrhizal (lanes
1-3) and Glomus intrøradices -colonised (lanes 4-6) roots of Hordeum vulgare cv.
'Galleon'. The extent of colonisation in the mycorrhizal roots was estimated to be
lSVo of the total root length.
The sizes of some of the DNA marker bands (Hae In digested ldv DNA) are shown
at the right hand side of the figure.
1234 5 6
2149bp
1097 bp
458 bp
FIGURE 5.2
Yield of total RNA extracted from 14 day old non-mycorrhizal (lanes 1-3) and
Glomus intraradices -colonised (lanes 4-6) roots of Hordeum vulgare cv. 'Galleon'.
The extent of colonisation in the mycorrhizal roots was estimated to be l87o of the
total root length. This graph represents the means of three samples for each treatment
of approximately equal weight and the standard errors of those means. Each sample
was extracted using a standardised proceedure.
Figure 5.2
0,3
bo
bo)
o
zú
0,28
0,26
0,24
0,22
o,2
NM M+
Treatment
FIGURE 5.3
Synthesis of first (lane l) and second strand (lane 2) cDNA reverse transcribed from
total RNA extracted from 14 day old Glomus intraradices -colonised Hordeum
vulgare cv.'Galleon' roots.
Aliquots were taken from the reaction mixes at the completion of each cDNA strand
synthesis and electrophoresed on a l.2Vo TAE agarose gel. The gels were wrapped in
'Gladwrap' plastic after the electrophoresis and exposed to X-ray film at room
temperature for 6 hours. The gels were then stained with ethidium bromide to
determine the position of the DNA markers (Hin¿ Itr digested I,DNA).
The approximate positions of some of the DNA markers are shown at the right of the
figure.
21
+ 4467 bp
? l974bp
+ 580bp
FIGURE 5.4
DNA sequence from the Imm 434 region of the phage vector l,gt10 and the two
primers used to amplify cDNA inserts by the polymerase chain reaction. The Eco
Rl cloning site is in bold print and the primer sequences (which anneal to the
complementary strands) are in bold print and underlined. The arrows indicate the
direction of theTaq polymerase-mediated DNA synthesis.
3
Forward (23mer) Primer
5I - CTTTTGAGCÀÀG'ITTCjÊ,GCCTGGTTå.EGTCCAAGCTGAÀTTCTTTTGCTTTTACCCTGGzu\GzuU\TACTCATAAGCCACCTCT - 3I
_ GA'UU\CTCGTTCA.A,GTCGGACCAA,TTCAGGTTCGACTTÀ.ÈGAÀÄACGAÃÃATGGGACCTTCTTTå,TGAGTA'ITCGGTGGAGA _ 5 '
Reverse (24mer) Primer
Figure 5.4
FIGURE 5.5
Gel electrophoresis of polymerase chain reaction amplified inserts from 15 phage
randomly selected from a cDNA library prepared from reverse transcribed total RNA
extracted from 14 day old Glomus intraradices -colonised Hordeum vulgare cv.
'Galleon'roots.
The DNA ma¡kers were Rs¿ llDra 1 digested pTZISU (left hand side) and Sau 3A
digested pTZlSU (right hand side). The sizes of two of the marker fragments are
shown at the right hand side of the figure.
<- 2l49bp
<- L42bp
108
5.2.3 Isolation of Putative Barley-mycorrhiza Related cDNA Clones
In view of the much better quality of the first cDNA library, it was used for the
initial differential hybridisation. Approximately 2.5 x 104 clones were plated out
(2,500 pfu/plate) and transferred (in duplicate) to Hybond N+ hybridisation
membranes (2.7.3.7). Radio-labelled first strand cDNA was prepared from total
RNA (25pÐ, extracted from the first 10 lateral roots of 14 day old NM and M+
barley roots, using Superscript H- reverse transcriptase (Bioprobe). One set of
membranes was then hybridised with NM-root derived cDNA and the other with
M*-root derived cDNA. After 16 hours of hybridisation, the membranes were
stringently washed (see 2.7.3.7) and then autoradiographed for 7 days at -80"C with
intensifying screens. The autoradiographs were developed and the hybridisation
patterns of duplicate membranes compared to identify clones with differential signal
intensities (see Figure 5.6). The plaques of 512 putative barley-mycorrhiza-related
(BMR) clones were identified in this manner and cored out with a Pasteur pipette,
expelled into 100p1 of SM buffer and allowed to diffuse into the buffer for 12 hours
at 4"C. Each clone was then numbered.
The membranes were then stripped of the remaining probe (Maniatis et aI., 1982)
and each set of membranes was re-hybridised with radio-labelled first strand cDNA
prepared from total RNA (25ttÐ extracted from either 14 day old NM or 14 day old
M+ barley leaves. This hybridisation was carried out because systemic responses to
pathogenic symbioses have been reported in plants (see Ryals et al., 1994). No
differential hybridisation signals were observed in the cDNA population analysed
and (not surprisingly) nearly 9OVo of the clones did not hybridise with the reverse
transcribed shoot RNA transcripts. It was concluded that most of the cDNA clones in
the library were homologous to root-specific mRNA species and that none of them
were detectable as being systemically induced in barley leaves in response to the
mycorrhizal infection process.
r09
Top agarose bacterial lawns were overlaid on LB medium in 135mm Petri dishes (as
per 2.7.2.3 but without adding any infective phage) and incubated for one hour at
37)C before lpl of the SM buffer containing each putative BMR clone was spotted
onto the plates in a regular pattern using a lcm grid. The phage were then allowed to
mutiply and lyse for 12 hours. These plaques were transferred to Hybond N+ in
duplicate and differential hybridisations were performed as described above except
that 40pg of freshly extracted plant total RNA was used for each of the reverse
transcription reactions.
The hybridisation patterns of the duplicate membranes (for an example see Figure
5.7) indicated that the majority of the putative BMR clones were false positives but
variations in signal intensity were observable in l2I of the putative BMR clones so
these were collected and renumbered; the remaining clones were discarded. None of
the clones displayed hybridisation patterns consistent with specific expression of
genes during mycorrhiza development but both increased and decreased
accumulation of specific RNA transcripts were observed. Both sets of membranes
were then stripped of the probe (Maniatis et al., 1982) and re-hybridised with radio-
labelled first strand cDNA reverse transcribed from shoot RNA as described in the
previous paragraph. The results were very similar to those described above; over
95Vo of the cDNA clones represented RNA species which are specific to plant roots
and none appeared to be differentially expressed.
The putative BMR clones remaining after the second round of differential screening
were then used to generate DNA probes by PCR amplification of the phage inserts as
previously described. Unfortunately, it was found that nearly 507o of the phage
samples generated mutiple bands, indicating a mixture of phage paficles (see Figure
5.8). Many of these mixed phage samples were replated on to 90mm petri dishes at
low plaque formation densities (100-200 pfu/plate) and six plaques from each plate
were isolated and checked by PCR amplification of the inserts. All samples
110
containing single inserts obtained in this manner were numbered and stored for later
use as DNA probes.
FIGURE 5.6
Identification of putative barley-mycorrhiza related cDNA clones by differential
screening of the mycorrhizal root cDNA library.
Phage DNA (approximately 25,000 plaques in all) was transferred, in duplicate, onto
two sets of Hybond N+ nylon membranes. One set of membranes were then
hybridised with 32-P-labelled single stranded cDNA prepared from non-mycorrhizal
(NM) root total RNA (25pg) whilst the other set of membranes were hybridised with
32-P-lab"Iled single stranded cDNA prepared ftom Glomus intraradic¿s -colonised
(M) root total RNA (25pg). The membranes ,were washed and then autoradiographed
at -80oC for 8 days with the aid of intensifying screens.
By comparing the hybridisation signal intensites of matched pairs of membranes, it
was possible to identify 512 cDNA clones that appeared to increase or decrease in
abundance during the earliest developmental stages of VA mycorrhiza formation.
These cDNA clones, which correspond to genes which could be either ulþSulateO À
(u) or down-regulated (d), are indicated by the arrows and were isolated and used for
a secondary differential screening.
FIGURE 5.6
u
MYC u N\,I
FIGURE 5.7
Secondary differential screening of the putative barley mycorrhiza related (BMR)
clones identified during the primary screening.
The putative BMR phage-clones isolated during the primary screening process were
used to tranfect a bacterial lawn and form large plaques (0.7-1.0cm). The phage
DNA was transferred to Hybond N+ nylon membranes, in duplicate, and each set of
membranes was hybridised with either 32-p-tabelted single stranded cDNA prepared
from non-mycorrhizal (NM) root total RNA (40pg) or 32-p-labelled single stranded
cDNA prepared from Glomus intraradic¿s -colonised (M+) root total RNA (25pg).
The membranes were washed and autoradiographed at -80oC for 6-8 days with
intensifying screens.
Putative BMR clones which are either upregulated (u) or down-regulated (d) are
indicated by the arrows.
HINI+Hi
tcftr'0
o
o
o
cr üÈls
ilaoö
at rlol ü I olf,oo o ö t i#Ð
o
to
I
t*
Ð
*
t
lo
t
t} rüI*
o Ðffi9Ð
oo
ü etrt I
t II
L '''+
tn
¿'s gunÐrd
FIGURE 5.8
Gel electrophoresis of polymerase chain reaction amplified inserts from 40 phage
randomly selected from the pool of putative barley-mycorrhiza related clones.
The clones containing multiple bands were replated at low densities of plaque-
formation (50-100 pfn/plate) and rescreened to ensure only single bands were present
in each phage preparation. All unique bands from each rescreening were kept for
analysis.
The DNA marker was Hind 111 digested lambda DNA and can be seen at the left
hand side of the gels. The sizes of two of the marker fragments are shown at the right
hand side of the figure.
+ l973bp
# 580bp
+ L973bp
? 580 bp
t14
5.2.4 DNA:RNA (Northern) Hybridisation Analysis
of Putative BMR cDNA Clones
The single band cDNA clones were used to compare the accumulation and
abundance of their respective homologous RNA transcripts in total RNA samples
extracted from the top half of whole root-systems of 7 and 14 day old NM and M+
barley roots and 14 day old M+ barley shoots by DNA:RNA (Northern)
hybridisation. The extent of total infection in the mycorrhizal barley plants was
calculated to be OVo and 21Vo in the 7 and 14 day old plants respectively (see 2.5).
These same plants were also used for the 14 day old M+ shoot RNA extractions.
Lateral roots were not specifically used for the preparation of these RNA membranes
because non-denaturing electrophoretic separations of these samples indicated that
slightly greater degradation occurred in lateral root RNA samples than whole root
samples, probably reflecting the longer time required for the root excision/extraction
process prior to the addition of CsCl to preparations before the overnight
centrifugation.
The five RNA samples (10¡rg/sample; 4 gels at a time) plus marker DNA were
electrophoresed under denaturing conditions (2.7.2.4) and transfered to Hybond N+
hybridisation membrane (2.7.2.5). These membranes were then hybridised to PCR-
generated putative BMR probes (2.7.4.1) which had been isolated and purified
(2.7.3.9) from a l.5%o agarose TAE buffered electrophoresis gel (2.7.3.6). The
approximate sizes of the hybridising RNA transcripts were determined by comparing
the migration of the hybridising band with a DNA size marker. A total of fifty six
hybridisations were carried out and each membrane was used for a maximum of
three hybridisations. The ribosomal RNA bands on the membranes were stained with
methylene blue (3Vo) in acetic acid (50Vo) after they had been used for the final time.
This allowed an estimation of the amounts of RNA that had been transferred to the
membrane and acted as a further aid to transcript length determination. New
115
membranes were always prepared from fresh extracts of total RNA. The mycorrhizal
plants always appeared uninfected after 7 days growth whilst the extent of infection
ranged from l5-22Vo after 14 days growth when I5Vo G. intraradices inoculum was
used. Freshly extracted RNA was used for each set of hybridisations to guard against
artefacts caused by contamination to an individual plant and to ensure
reproducibility of the results. The RNA extracted from the 7 day old material
(uninfected in both NM and M+ plants) served two purposes; it was a fuither control
against contamination-induced changes in one group of plants; and it allowed the
possibility of detecting transcripts of genes which are up- or down-regulated prior to
or at the very early (undetectable) stages of mycorrhiza development. The leaf total
RNA served as a negative control to identify cross hybridisation (particularly with
ribosomal RNA) and background hybridisation. It may also have identified systemic
responses not detected during the differential screening process.
Forty one of the cDNA probes used for the Northern analyses detected only very
slight differences in the accumulation of the homologous RNA transcripts between
the NM or Mr roots. These differences were most probably due to slightly uneven
loading or transfer of the gels (the large number of false positives obtained all
through the screening process highlights some of the difficulties involved in using
untargeted cloning strategies). No obvious differences in transcript accumulation
were detected by any of the cDNA probes between NM and M+ 7 day old root RNA
samples. Seven of the clones hybridised very weakly with the total root RNA; the
transcript levels of those particularly RNA species appeared to be at the limit of
detection.
One cDNA probe (BMR37) detected a root-specific 1100b RNA transcript which
accumulated to much higher levels in 14 day old roots of NM or M* plant than in 7
day old roots of NM or M+ plants (Figure 5.9). Two cDNA probes (BMR64 and
BMR96) detected root-specific RNA transcripts (1300b and 200b in length,
116
respectively). rWhile equivalent in abundance in J day old roots of either NM and
M+ plants these sequences appeared to much less abundant in 14 day old M+ roots
than in NM roots of the same age (Figure 5.9).
Finally, two probes (BMR6 and BMR78) detected RNA transcripts (1100b and
3500b respectively) which were more abundant in the M+ 14 day old roots than in
NM roots of a corresponding age (Figures 5.10 and 5.11). The genes which encode
these RNA species appear to be constitulely expressed at a low level (7-10 dayA'r
exposure times were required for the autoradiographs) but are up-regulated during
the formation of mycorrhizas. It appears that they may be actively involved in the
response to or formation of VA mycorrhizas. However, it should be noted that the
increased expression of the genes detected by BMR6 and BMR78 correlates with the
onset (and as shown for BMR6, the extent) of dramatic fungal development. That
these two cDNA clones are of fungal or even bacterial origin (see review on
mycorrhizal helper bacteria by Garbaye; 1993) cannot be ruled out. It is unlikely but
possible that the constitutive levels of expression observed in the young barley roots
were caused by cross-hybridisation of both these cDNA clones with a barley-root
RNA of approximately the same size as a fungal or bacterial RNA. To ensure that
the clones were monitoring the increased accumulation of specific plant mRNA
species and not simply fungal development, further experiments were undertaken to
confirm that the cDNA clones, BMR6 and BMR78, were of plant origin.
5.2.5 DNA:DNA (Southern) Hybridisation Analyses
of the two cDNA Clones BMR6 and BMR78.
High molecular weight genomic DNA was extracted from leaves of H. vulgare cv.
'Galleon' leaves, H. vulgare cv. 'Betzes' leaves (2.7.3.I),21 day old heavily
mycorrhizal (4O-50Vo) 'Galleon' roots, which were grown as previously described,
and sterilised G. intraradices spores (as per 2.2.2) which were kindly supplied by A.
II7
Tassie. The plant leaves and mycorrhizal roots yielded 50-70¡rg of DNA, the quality
of DNA extracted, as determined by minigel electrophoresis, was excellent. The G.
intraradices spores yielded 7-8pg of what appeared to be good quality, high
molecular weight genomic DNA.
A sample of DNA from the three plant sources (8ttg) and all the G. intraradices
DNA was digested with the restriction enzyme Eco Rl. Each DNA digestion
reaction was divided into two equal samples and the fragments were separated by
overnight gel (1 .5Vo TAE agarose) electrophoresis (2.7.3.6). The DNA fragments
were transferred to Hybond N+ membranes and hybridised (2.7.3.7) with
radiolabelled PCR-generated insert (2.7.4.1) from either BMR6 or BMR78 before
washing and exposure to X-ray film at 80oC for three days. The results of the
hybridisations are shown in Figure 5.11.
Both probes hybridised to the Eco RI digested DNA extracted from the two barley
cultivars and the mycorrhizal 'Galleon' roots. BM6 detected two bands (of
approximately 4.0 kbp and 5.0 kbp in size) in the 'Galleon' samples and a single
band of approximately 4.0 kbp in 'Betzes'. BMR78 detected a single band of
approximately 3-3.5 kbp in the two 'Galleon' samples and a slightly larger in
'Betzes'; this may have been an a¡tefact caused by the gel. The hybridisation signals
were weaker for 'Betzes' DNA than both the 'Galleon' DNA samples for both the
probes, indicating that the loading of the wells may not have been even. No signal
was detetable in the fungal spore DNA lane (which almost certainly has a much
smaller genome), even after 10 days exposure (not shown). Subsequent reprobing of
the membranes with radiolabelled fungal DNA showed that the fungal DNA had
been efficiently transferred to the membranes (not shown). It is likely that the cDNA
clones were derived from plant RNA's and are not of fungal or bacterial origin.
FIGURE 5.9
Autoradiographs of DNA:RNA hybridisations using 32-p-labelled inserts from the
putative barley-mycorrhiza related clones BMR37, BMR64 and BMR93 to probe
total RNA extracted from the tissues of barley of differing ^ge
and mycorrhizal
status.
Total RNA (lOpg/lane), extracted from the roots of 7 day old (lane l) and 14 day old
(lane 2) Glomus intraradices -colonised Hordeum vulgare cv. 'Galleon' , 7 day old
(lane 3) and 14 day old (lane 4) non-mycorrhizal 'Galleon' and the shoots of 14 day
old (lane 5) myconhizal plants, was gel electrophoresed under denaturing conditions
and transferred to Hybond N+ membranes. The probes were generated using the
polymerase chain reaction and labelled by random priming. The membranes probed
with BMR37, BMR64 and BMR96 were autoradiographed at -80oC with
intensifying screens for 8, 5 andT days respectively.
The approximate sizes of the hybridising RNA bands, shown at the right of the
figure, were determined by comparing the distances of migration of the hybridising
bands with those of the 28S and 18S ribosomal RNA bands.
The lane order for the membrane probed with BMR93 has been rearranged to match
the other two autoradiographs.
FIGURE 5.9
BMR37
MNM234
BMR64
M NM S234 5
+- 1.1 kb
+- 1.3 kb
+- 2.0 kb
s5
NM2
M43
1
s5
I
ËË
BMR93
FIGURE 5.10
Autoradiographs of DNA:RNA hybridisations using 32-P-labelled insert from the
putative barley mycorrhiza related clone BMR6 to probe total RNA extracted from
the tissues of barley of differing age and mycorrhizal status.
A: Total RNA (lOp gllane), extracted from the roots of 7 day old (lane 1) and 14 day
old (lane 2) Glomus intraradices -colonised Hordeum vulgare cv. 'Galleon' ,7 day
old (lane 3) and 14 day old (lane 4) non-mycorrhizal 'Galleon' and the shoots of 14
day old (lane 5) mycorrhizal 'Galleon', was size fractionated by gel electrophoresis
(under denaturing conditions) and transferred to Hybond N+ membranes. The phage
cDNA insert was generated using the polymerase chain reaction (PCR) and labelled
by random priming using the Klenow fragment of DNA polymerase 1. The
membranes were autoradiographed for 7 days at -80oC with intensifying screens.
B: Total RNA (l5pg/lane) was extracted from the roots or shoots of 14 day old
Hordeum vulgare cv. 'Galleon' plants with varying extents of. Glomus intraradices
colonisation: roots which were IOVI colonised (lane 1); roots which were 25Vo
colonised (lane 2); roots which were 30Vo colonised (lane 3); roots which were not
colonised (lane 4); and shoots from plants with roots which were 25Vo colonised
(lane 5). The RNA (lOpg/lane) was size fractionated by gel electrophoresis (under
denaturing conditions) and transferred to Hybond N+ membranes. Phage cDNA
insert was PCR amplified and labelled as above. The membranes were exposed for 8
days at -80oC with intensifying screens
The approximate sizes of the hybridising RNA bands, shown at the right of the
figure, were determined by comparing the distances of migration of the hybridising
bands with those of the 28S and 18S ribosomal RNA bands.
FIGURE 5.10
AM NM S23 4 5
BMR6
B
r234 5
BMR6
{- 1.3 kb
<- 1.3 kb
I
FIGURE 5.11
Autoradiograph of DNA:RNA hybridisation using 32-P-labelled insert from the
putative barley mycorrhiza related clone BMR78 to probe total RNA extracted from
the tissues of barley of differing age and mycorrhizal status.
Total RNA (lOpg/lane), extracted from the roots of 7 day old (lane l) and 14 day old
(lane 2) Glomus intraradices -colonised Hordeum vulgare cv. 'Galleon', 7 day old
(lane 3) and 14 day old (lane 4) non-mycorrhizal 'Galleon' and the shoots of 14 day
old (lane 5) mycorrhizal 'Galleon', was gel electrophoresed under denaturing
conditions and transferred to Hybond N+ membranes. The phage cDNA insert was
generated using the polymerase chain reaction and labelled by random priming using
the Klenow fiagment of DNA polymerase 1. The membrane was autoradiographed
for 5 days at -80"C with intensifying screens.
The approximate sizes of the hybridising RNA bands, shown at the right of the
figure, were determined by comparing the distances of migration of the hybridising
band with those of the 28S and 18S ribosomal RNA bands.
FIGURE 5.TI.
M NM S234 s
BMR78
<- 3.2 kb
1
FIGURE 5.12
Autoradiograph of DNA:DNA hybridisations using 32-P-lab"lled inserts from the
putative barley mycorrhiza related clones BMR6 and BMR78 to probe genomic
DNA from two barley cultivars and the mycorrhizal fungus, Glomus intraradices.
High molecular weight DNA (4-5pg), extracted from the shoots (lane 1) and Glomus
intraradices -colonised roots (lane 2) of Hordeumvulgare cv 'Galleon', the shoots
(lane 3) of H. vulgare cv. 'Betzes' and spores (lane 4) of the mycorrhizal fungus
Glomus intraradice.s , was digested with Eco Rl and the fragments size fractionated
by gel electrophoresis. The DNA was transferred to Hybond N+ membranes and
hybridised with either BMR6 or BMR78 which had been labelled by random
priming using the Klenow fragment of DNA polymerase 1. The membranes were
then autoradiographed for three days with intensifying screens.
Relevant DNA size markers are shown at the right hand side of the figure.
!II!I
I¡
I
¡ii
L
t2r
5.3 Discussion
The approach taken to identify BMR genes was to differentially screen a cDNA
library prepared from G. intraradices colonised barley roots. This sucessfully
identified a number of interesting clones that fell into several categories based upon
patterns of gene-expression associated with colonisation of the roots by the
mycorrhizal fungus.
One clone, BMR37, detected a gene that showed increased expression in older roots.
Seven day old 'Galleon' root systems are much smaller than fourteen day old
'Galleon' roots and have already developed many lateral root primordia (see Chapter
3). As such, the seven day old root systems may have a much greater proportion of
active root tip cells to the metabolically less active transport cells distal to the
growing tips than fourteen day old root systems. The increase in expression of the
gene which encodes the BMR37 RNA-homologue in fourteen day old root systems,
may reflect an increase in the proportion of cells (for example the distal transport
cells) expressing this gene as the root system ages and expands. Spatial analysis of
the transcript distribution of the BMR37 RNA-homologue within the root by either
Northern or in situ hybridisation methods or PCR-based detection should answer
this question.
The observed decreases in abundance of RNA transcripts in M+ root material
detected by BMR64 and BMR96 are not easily explained. Similar expression
patterns have been detected by Pisolithus tinctoris cDNA clones during
ectomycorrhiza formation with Eucalyptus globulus roots (Tagu and Martin, 1993).
The down regulation of the genes from which the RNA's are transcribed appears to
be related to the differentiation of the fungus from the free-living state to the
symbiotic state (Murphy and Tagu, unpublished). At the cellula¡ level, root cells also
undergo considerable cytological change during mycorrhizal development (see my
122
comments in Chapter 1) particularly in cells containing arbuscules. These changes
appear to involve metabolic activation of the plant cells; for example, the synthesis
of membranes and (presumably) trans-membrane transport machinery.
It is tempting to speculate that the RNA transcripts detected by these two cDNA
probes encode regulatory gene products: The mycorrhizal fungus could directly or
indirectly repress these regulatory genes, which in turn derepress other genes, the
translation products of which are necessary for functional mycorrhiza establishment.
Such an elaborate regulatory system is not inconceivable in what appears to be (in
evolutionary terms) an extremely old symbiotic association. The possibility that they
detect mRNA's which encode proteins involved in plant defence systems that are
repressed during mycorrhiza development or induced in the non-mycorrhizal roots
because of contaminating micro-organisms, cannot be ruled out. The Northern
analyses indicated that both mRNAs accumulated at low to moderate levels in non-
mycorrhizal roots. This may reflect gene expression induced by the harvesting
process. Pathogenic infection and wounding can result in the rapid accumulation of
specific RNA species, particularly those encoding enzymes of phenolpropanoid
biosynthesis (Lamb et al., 1989), and protein-synthesising capacity (Davies and
Schuster, 1981). Sequencing these cDNA clones could very quickly confirm this
hypothesis as defense-related genes are relatively well characterised and many DNA
sequences are now available (see Dixon and Lamb, 1990; Bowles, 1990). This
should enable a DNA data base search to establish the identities of cDNA's with
homologies to any of the well characterised defense-related genes.
BMR6 and BMR78 detected the RNA transcripts of genes which are constituvely
expressed at a low level (5-8 day exposure times were required for the
autoradiographs) and upregulated during the formation of mycorrhizas. The genes
homologous to these cDNA's appear to be actively involved in the formation of, or
response to, VA mycorrhiza development. These two clones were analysed in much
123
greater detail in order to ascribe a putative function to the translation products.
Chapter 6 describes the preliminary molecular characterisation of the two cDNA
clones, BMR6 and BMR78.
t24
r
Molecular Characterisation of the
Barley Mycorrhiza-related cDNA's :
BMR6 and BMR78
6.L Introduction
The cDNA clones, BMR6 and BMR78 were shown to detect mRNA transcripts which
increase in abundance during the early stages of mycorrhiza development in Glomus
intraradices infected Hordeum vulgare cv. 'Galleon' roots (see Chapter 5). The aims
of the series of experiments described in this chapter were to characterise these clones
in more detail and, to ascribe a putative biological function to the products of the genes
represented by these cDNA clones. Both clones were subcloned into the plasmid,
pTZISU, and used as probes in further DNA:RNA (Northern) hybridisation
experiments to determine whether the increased accumulation of the corresponding
mRNA species was specific to the development of myconhizal roots or was a general
response induced by pathogens and/or water stress. It was also important to determine
if the accumulation of mRNA transcripts was influenced by the phosphate nutrition of
the non-mycorrhizal barley plants, because of the important influence of mycorrhizal
colonisation on phosphate uptake from the soil (see Chapter 1). The clones were also
used to map the positions of the corresponding genes to the barley genome using
DNA:DNA (Southern) hybridisation analysis of barley addition lines and doubled
haploid mapping populations. Finally, the sub-clones were sequenced and the
sequences compared to available DNA and protein sequence data banks because high
degrees of homology between the sequences of BMR6 and BMR78 and previously
characterised genes could provide some strong evidence of their biological functions.
t25
6.2 Subcloning of BMR6 and BMR78
into the Plasmid Vector pTZ18U
Insert (approximately 250ng) obtained from large scale phage preparations of BMR6
^ala nVfRZS was ligated into the Eco Rl restriction site of the mutifunctional plasmid
vector, pTZlSU which is based on the vector, pUC18 (see Yanisch-Perron, 1985).
The ligation products were used to transform competant Eschereshia coli DH5a (see
2.7 .L.3) which were then spread onto LB agar plates containing ampicillin (75mglml),
X-gal and ITPG (see Maniatis et aI., 1982).
Twelve white colonies from the BMR6 transformation and all 5 white colonies from
the BMR78 transformation were picked from the plates and used to prepare plasmid
minipreparations (2.7.3.2). The insert size of the isolated plasmids was checked by
Eco RI digestion and visualisation of the restriction fragments after separation by gel
electrophoresis (2.7.3.6). All of the BMR6 plasmids contained inserts of
approximately 650bp which matched the size of the original phage insert. All of the
BMR78 transformants contained plasmid DNA but only one contained an insert that
was recoverable by Eco Rl digestion. The insert size of this clone matched that of the
original phage insert (approximately 580bp). Subsequent re-culturing of this clone,
both on LB plates and in LB liquid medium containing ampicillin (75¡tglml) indicated
that the pBMR78 insert was unstable in E coli DH5a. Consequently, further ligations
and transformations were carried out (as above) using E. coli strains HB101 and
ER1647: Strain ER1647 yielded 5 stable pBMR78 clones.
The restriction maps prepared for each clone (see Zehetner, 1987) have been
incorporated into the sequencing-strategy diagram (see Figure 6.6).
t26
6.3 Expression of the BMR6 and BMR78 Genes
6.3.1 Aims
The aim of the experiment described in this section was to grow plants under a variety
of conditions in order to test whether factors other than mycorrhizal colonisation
resulted in the increased accumulation of the mRNA species detected by BMR6 and
BMR78.
Two biovars, Ggt-800 and Ggg-V/2P, of. the root fungal pathogen Gaeumannomyces
graminis (often known as 'Take-all') were used to colonise barley seedlings for one
part of this experiment. Both strains of G. graminis colonise the epidermal and
cortical cells of the root but Ggt-800 can also invade the stele. Consequently Ggt-800
is virulent to many barley cultivars because it blocks the vascular tissues resulting in a
marked reduction in the growth of the above ground parts (particularly the seed) of the
plant. Ggg-W2P is considered avirulent with barley but colonises the outer root cells
as much or even more (as judged by root blackening) than Ggf800 (Harvey, 1994).
Another group of barley seedlings were subjected to water-stressing as an example of
environmental stress. Increased accumulation of the transcripts detected by BMR6
and/or BMR78 under any or all of the treatments used in this experiment, would
suggest that the genes corresponding to the two cDNA's are up-regulated by
conditions more general than just mycorrhizal development. As such they could be
related to a large group of genes which are known to be induced as a general response
to pathogen colonisation and/or environmental stresses such as drought and wounding
by feeding insects or herbivores. This group includes the genes that encode the PR,
chaperonin and polyphenolpropanoid pathway proteins (see Dixon and Lamb, 1990;
Linthorst, 1992; and Ellis, 1990).
Barley seedlings were also grown under different phosphate regimes to see ifphosphate availability (which is probably largely responsible for the mycorrhizal
growth responses described in Chapter 3) induced increased accumulation of these
r27
RNA transcripts. The transcripts may be produced from phosphorous metabolism-
associated rather than mycorrhiza-development genes.
Non-myconhizal and mycorrhizal control plants, without any added phosphate, were
grown as previously described in Chapter 2. A list of the treatments used in this
experiment and the number of replicates for each treatment are shown in Table 6.1.
Table 6.1
List of treatments for the experiment designed to assess the effect of infection byvirulent and avirulent 'Take-All', phosphate nutrition and water stress on the
expression of the genes corresponding to the cDNA clones BMR6 and BMR78.
6.3.2 Plant Growth
The G. graminis inoculations of plants were carried out by burying a 1cm3 agarose
block excised from a petri dish covered with actively growing fungus of either
biotype, approximately 5cm below the soil medium surface in each treatment pot (see
Riggleman et al., 1985) five days before transplanting of the barley seedlings. The
plants were ha¡vested 14 days after transplantation. The extent of infection in the G.
graminis infected plants was not quantified, but extensive black discolouration (over
5OVo of the root length), which is associated with 'Take-all' infection, was clearly
discernible on all of the root systems that were ha¡vested. The shoots of all the plants
appeared to be normal when visually compared with NM:Pg and M+:Pg control plants
(see Chapter 3). Total RNA was then extracted from the top half of those root systems
(2.7.2.1) with extensive colonisation by either the virulent and avirulent G. graminis
Treatments Replicates
Non-mycorrhizal with no added phosphate (NM:Po)
Non-mycorrhizal with added phosphate (0.5 mmoles/kg; NM:P1)
Mycorrhizal with no added phosphate (M+:Po)
NM with phosphate added (0.5 mmoles/kg) after 12 days (NM:Pa¿¿)
NM:P6 and water-stressed
NM:Pg and grown with virulentG.graminis (Ggt-800)
NM:Po and grown with avirulentG.graminis (Gee-W2P)
5
5
5
5
25
5
5
r28
biotypes.
Twenty plants were used in the water-stress component of this experiment. All plants
were set up as previously described for NM:P9 treatments (see Chapter 3). 'Water-
stress was applied by discontinuing daily watering to five randomly selected plants
after 71819 or 10 days growth. These plants were visually compared with the control
plants (see above) which were watered to weight (l2%o wlv) every day throughout the
experimental period. All of the unwatered plants became visibly water-stressed within
6 days of the discontinuation of the daily watering whilst all the control plants
appeared normal at 14 days. The five plants which did not receive additional watering
after 8 days growth (and had been wilted for approximately 12 hours) were harvested
and the total RNA was extracted from the top half of each root system as previously
described (2.7.2.1).
There were 4 phosphate treatments (again see Table 6.1). The NM:P4¿¿ treatment
(plants without added phosphorous but receiving phosphate in solution after 12 days
growth) was an attempt to 'mimic' the changes in phosphate physiology within the
root which may be brought about by a sudden increase in phosphate availability due to
the rapid and extensive formation of a functional mycorrhizal root system. All plants
were harvested after 14 days growth and total RNA was extracted from the top half of
each root system and the shoots of the M+:P6 plants.
6.3.3 Expression Analysis
Two sets of hybridisation membranes were prepared using total RNA (l5pg/lane)
isolated from plants receiving each of the treaments described above.
These were then hybridised (2.7.2.6) with radio-labelled DNA probes prepared from
either pBMR6-1 or pBMR78-l (2.7.3.11).
r29
6.3.4 Results
The results of the Northern hybridisation experiments are shown at Figures 6.1 and
6.2. Figure 6.1 shows that the accumulation of RNA transcrþts detected by pBM6-1
and pBM78 (after 7 days exposure of the membranes) was higher in the roots of the
mycorrhizal control plants than in the non-mycorrhizal controls. This agrees with the
results previously obtained (see Chapter 5). The accumulation of these transcripts was
not increased by infection with either of the G. graminis biotypes. It was concluded
that the genes corresponding to BMR6 and BMR78 a¡e unlikely to belong to the group
of genes which are induced by many pathogens and environmental stresses, although
water-stressing appeared to increase accumulation of both the BMR6-I and BMR78
transcripts to a small degree. This does not appear to be a result of uneven loading
because the ethidium bromide staining of the original gels and the methylene blue
staining of the RNA after use of these two membranes (data not shown) showed no
obvious differences between lanes. The apparent increase in expression of both clones
in the water stressed roots may reflect changes in the ratio of the mRNA to ribosomal
RNA fractions (which is nearly all of the total RNA) brought about by dehydration of
the roots. As water-stress induction of the corresponding genes cannot be ruled out,
more detailed investigations are warranted.
The hybridisation membranes prepared with RNA extracted from the phosphate
treatments were exposed for 10 days. These membranes did not contain a M+:Pg
control lane (the RNA sample was degraded) so direct comparisons between them and
the membranes prepared with RNA from the water-stressed or pathogen-colonised
roots are not possible. However, it is clear that the accumulation of BMR6 and
BMR78 detectable RNA transcripts was unaffected by the three phosphate treatments
used (see Figure 6.2). The results give no support to the idea that increased
accumulation of either transcript is a response to increased phosphate nutrition, rather
than colonisation by the fungus itself.
130
It was concluded that the gene expression patterns observed in this experiment for
BMR6 and BMR78, support the hypothesis that the genes corresponding to these two
cDNA clones are up-regulated during mycorrhiza development. Further, these genes
do not appear to be involved in the general plant defence responses to pathogen-
colonisation and/or environmental stresses or in phosphate uptake and metabolism.
FIGURE 6.I
Expression analysis of BMR6 and BMR78: DNA:RNA hybridisations comparing the
accumulation of mRNA transcripts in the roots or shoots of 14 day old Hordeum
vulgare cv. 'Galleon' plants grown under various conditions.
Total RNA (l5pg/lane), extracted from non-mycomhizal roots (lane l), mycorrhizal
roots (lane 2), avirulenlGaeumannomyces graminis - infected roots (lane 3), virulent
G graminis - infected roots (lane 4), water-stressed roots (lane 5), and the shoots of
mycorrhizal plants (lane 6), was size fractionated by gel electrophoresis under
denaturing conditions and transferred to Hybond N+ membranes. Insert DNA was
isolated from pBMRó-1 and pBMR78-1 and labelled by random priming using rhe
Klenow fragment of DNA polymerase L The membranes were autoradiographed for 7
days at -80'C with intensifying screens.
The approximâte sizes of the hybridising bands, shown at the right of the figure, were
determined by comparing the distances of migration of the hybridising bands with
those of the 28S and 18S ribosomal bands.
BM78L 2 3 4 s 6
BM6
12 3 4 5 6
<-- 3.5 kb
F 1.3kb
FTGURE 6.2
lnfluence of phosphate nutrition on the accumulation of mRNA transcripts detected by
BMR6 and BMR78 in non-mycorrhizal roots of 14 day old Hordeum vulgare cv.
'Galleon'.
Total RNA (15pg/lane), extracted from the roots of plants grown without additional
phosphate (Ps), grown with additional phosphate (0.5mmoles/kg) (Pt)' and grown
without additional phosphate but receiving phosphate after 12 days (0.5 mmoles/kg),
was size fractionated by gel electrophoresis under denaturing conditions and
transferred to Hybond N+ membranes. Insert DNA was isolated from pBMR6-1 and
pBMR78-1 and labelled by random priming using the Klenow fragment of DNA
polymerase 1. The membranes were autoradiographed for 7 days at -80oC with
intensifying screens.
The approximate sizes of the hybridising bands, shown at the right of the figure, were
determined by comparing the distances of migration of the hybridising bands with
those of the 28S and 18S ribosomal bands.
Po Pr P6
Po Pr Pt
r33
6.4 Chromosomal Mapping of the
cDNA Clones BMR6 and BMR78.
6.4.r Aims
The aim of the series of experiments described in this section was to produce
chromosome maps of barley showing the genetic positions of the genes corresponding
to BMR6 and BMR78 in relation to other markers which had already been mapped.
One of the long term aims of the V/aite Institute cereal breeding program is to identify
molecular markers linked to phenotypic traits because it is envisaged that this could
reduce the time and cost involved in selecting breeding lines for further crossing
and"/or release.The genes encoding BMR6 and BMR78 were mapped as part of the
ongoing frrapping project and to further cha¡acterise these clones.
Three doubled haploid mapping populations had been generated by crossing the barley
cultivars Galleon x Haruna Nijo, Clipper x Sahara 377I and Chebec x Harrington
(Langridge et al., unpublished). Doubled haploids generated from these crosses
(which constitute the mapping populations) were used to determine linkage groups for
a large number of DNA sequences (marker probes); these marker probes were selected
because they had been shown to reveal restriction fragment length polymorphisms
(RFLP's), detectable by Southern hybridisation, between the parent cultivars of one or
more of the above sets of doubled haploid populations. The probes were assigned to
and ordered within linkage groups based on the segregation of these RFLP's in the
doubled haploids. The software program MAPMAKER was used for this analysis.
The ready availability of barley cv. 'Betzes' addition and ditelosomic lines (Islam er
a.1., 1981) also enabled the rapid assignment of many probes to a particular
chromosome arm.
6.4.2 Genetic Mapping
The first step in localising the genes corresponding to BMR6 and BMR78 was to
identify any RFLP's between the sets of parental lines which could be used for the
t34
segregation analysis of the respective doubled haploid mapping populations.
Accordingly, radio-labelled probe, prepared from either pBMR6-l and pBMR78-1
was hybridised to duplicate membranes containing total DNA (10 pg) extracted from
each of the parental lines digested with each of the following restriction endonuleases:
EcoRl, Dra I, Hind lll, Bam Hl and Eco RV.
Both clones detected a single band of moderate intensity in at least one of the
restriction endonuclease/barley cultivar combinations that were tested (data not
shown). BMR6 detected RFLP's in several of the combinations. The Bam Hl
digestion of Clipper and Sahara 377t genomic DNA produced an unambiguous RFLP
which was chosen for the linkage analysis. BMR78 only produced one RFLP, in the
Dra I digestion of 'Galleon' and Haruna Nijo. Fortunately, it was unambiguous and
was selected for further analysis.
Radio-labelled probe prepared from pBMR6 was then hybridised to membranes
containing BamHI digested genomic DNA from each of the doubled haploids of the
Clipper x Sahara 3771 cross. Similarly, radio-labelled probe prepared from pBMR78
was hybridised to Dra I digested total DNA extracted from each of the doubled
haploids recovered from the Galleon x Haruna nijo cross. The autoradiograms of the
membranes were then visually assessed to determine the segregation of the
polymorphic bands in each doubled haploid plant. These data were then analysed
using the MAPMAKER (Lander et a1.,1987) program.
The gene corresponding to the cDNA clone BMR78 appears to be located on the long
arm of barley chromosome 2H (see Figure 6.3). The BMR78 RFLP co-segregated
with many clones (not all are shown in Figure 6.3) that had been unambiguously
mapped, using the wheat/barley chromosome addition and ditelosomic lines, to the
long arm of chromosome 2. A similar chromosome 2H genetic map generated using
the doubled haploids obtained from the Clipper x Sahara 3771 cross contained several
markers (but not BMR78) which were coflrmon to both maps. A comparison of these
two maps showed that the markers common to both maps were in the same linear
135
order and that the genetic distances between these loci were in good agreement.
Moreover, BMR78 maps to a region of chromosome 2H which contains markers
generated by two other mycorrhizal-root cDNA's.
The chromosomal position obtained for BMR6 was less reliable than that obtained for
BMR78 because only two other probes were in the same linkage group as BMR6.
This was primarily because the mapping project is being ca:ried out in chromosome
number order (ie. 1H-7H) and, at the time of this investigation, more markers had
been mapped for chromosomes 1H-3H than chromosmes 4H-7H. Both of the markers
linked to BMR6 generated multiple RFLP's which map to various chromosomes.
However, one of the RFLP's generated by each of the linked markers do map to a
common region. On this basis the gene conesponding to BMR6 was placed onto the
long arm of chromosome 6H.
These data showed that the gene corresponding to BMR78 is almost certainly situated
on the long arm of barley chromosome 2H and strongly suggested that the gene
corresponding to BMR6 is located on the long arm of barley chromosome 6H. The
locations were further investigated using the wheat/barley-chromosome addition lines
generated by Islam, et aL, (1981). It was hoped that this would confirm the results
obtained from the mapping experiments.
6.4.3 Addition Line Mapping
Total genomic DNA was extracted (2.7.3.1) from 'Chinese Spring' wheat, 'Betzes'
barley and wheat barley/addition lines 1H-7H, digested with Hind III, and transfered
to Hybond N+ hybridisation membrane. The membranes were hybridised with radio-
labelled probe prepared from either pBMR6 or pBMR78.
The results of these hybridisations (see Figures 6.4 and 6.5) confirmed that the
genomic sequences homologous to BMR6 and BMR78 are situated on barley
chromosomes 6H, and 2}J, respectively. The signal intensities of the strongly
hybridising bands identified in this experiment and the mapping hybridisation
136
experiments indicated that these cDNA's represent the partial-sequences of single or
low copy-number genes. Not surprisingly, hybridising bands were detected in the
'Chinese Spring' wheat DNA by both probes and probably represent heterologous
genes present in this closely-related cereal species. Longer (12-t4 day) exposure of
each hydridisation membrane (not shown) identified several other bands in the
digested barley DNA which weakly hybridised to both probes, indicating that the gene
sequences detected by BMR6 and BMR78 may represent highly divergent multigene
families.
BMR78 also detected a band in addition line 3H which did not correspond to any of
the hybridising bands in 'Chinese Spring' or 'Betzes'. This band may represent the
large (7.5 kb), strongly hybridising band in 'Chinese Spring' which is common to all
addition lines but only weakly hybridising in addition line 3H. The polymorphism
could indicate contamination of the seed stock used for genomic DNA preparation or a
mutation in the 'Chinese Spring' genomic component of addition line 3H.
RCURE 6,3
Genetic linkage maps of Hordeum vulgare chromosomes 2H (6.3a) and 6H (6.3b)
showing the positions of BMR6 and BMR78.
The maps were generated by linkage analysis (using the MAPMAKER program) of
doubled haploid populations derived from the crossing of Galleon x Haruna Nijo (for
ch¡omosome2H) and Clipper x Sahara (for chromosome 6H).
Distance MarkerName
ksr¡D22
14.8
cDo373
8.0
BBarley Chromosome 6H
Distance Ma¡kerName
BMR6
PSRl19
29.0
ksuA3
cM
3.8
cM
1.1
'7 ','
1.0
5.5
1.0
11.5
3.4
0.92.9
9.7
BMR78cD680
ksr¡F15Bæ266
ksr¡F41ABC153
@o36
BCDÍ10wG645BCD339
AWBN4Al7
Figure 63
NGURE 6.4
Chromosome mapping of BMR6 using wheat/barley addition lines lH - 7H (Islam ør
al.,l98l).
Plant genomic DNA (10pg) rromTriticum aestivum cv.'chinese Spring' (cs),
Hordeum vulgare cv. 'Betzes' (B), and wheat/barley addition lines lH-7H (lanes
marked 1-7) was digested to completion with Hind III, size fractionated by gel
electrophoresis (17o agarose) and transferred to Hybond N+ hybridisation membranes.
The membranes were hybridised overnight with radiolabelled probes generated by
random-primer labelling of insert derived from phage BMR6 by polymerase chain
reaction amplification. The membranes were washed and autoradiographed for 7 days
at -80oC with intensiffing screens.
The hybridising bands at the far right of the figure are from the marker DNA, Hind
111 digested ÀDNA. The positions of the 6742bp and 580bp bands are indicated on
the right hand side of the figure.
csB 1 2 3 4 5 6 7 M
æ 6742bp
ts 580bp
;
ti
If
¡
tI
II
f
I
:
FIGURE 6.5
Genetic mapping of BMR78 using wheat/barley addition lines lH - 7H (Islam et al.,
198 t).
Plant genomic DNA (10trg) rromTriticum aestivum cv.'chinese spring' (cs),
Hordeum vulgare cv. 'Betzes' (B), and wheat/barley addition lines 1H-7H (lanes
marked l-7) was digested to completion with Hind III, size fracrionated by gel
electrophoresis (.lVo agarose) and transferred to Hybond N+ hybridisation membranes.
The membranes were hybridised overnight with radiolabelled probes generated by
random-primer labelling of pBMR78-l insert DNA. The membranes were washed and
autoradiographed for 7 days at -80oC with intensifying screens.
The positions of the 6742bp and 580bp bands of the marker DNA, Hind 11 l-digested
ÀDNA, a¡e indicated on the right hand side of the figure.
csB t 2 3 4 s 6 7
æ6742W
? L974bp
t40
6.5 Sequencing of the cDNA Clones BMR6 and BMR78
6.5.1 Introduction
The aims of these experiments were to determine the DNA sequences of the cDNA
clones BMR6 and BMR78. The DNA sequences and the putative protein sequences
deduced from the DNA sequences could then be compared to the existing data banks.
Regions of these clones with high or even moderate homology to other, better
characterised genes, could provide some evidence of the biological function of the
genes which correspond to BMR6 and BMR78.
6.5.2 Sequencing Reactions
The phage inserts, BMR6 and BMR78, were cloned into the plasmid vector pTZlSU
(see section 6.1) primarily for ease of handling (ie. DNA replication, isolation and
storage) but also because it is suitable for PCR-assisted sequencing (based on the di-
deoxy method of Sanger et al., 1977) using commercially available primers. The
restriction maps and strategies used for sequencing (2.7.3.13) each clone are shown in
Figure 6.6 and the deduced DNA sequences are shown in Figure 6.7.
6.s.3 Sequence Analyses of BMR6 and BMR78
The BLAST program (Altschul et aL, 1990; Gish and States, 1993) was used to
compare the DNA and predicted protein sequences of BMR6 and BMR78 with
existing data bases containing reported DNA and protein sequences. By comparing
these sequences with known sequences, it was hoped that putative biological functions
could be ascribed to each ofthese clones.
The first 250 bases of BMR78 proved to be highly homologous (over 80Vo) at the
DNA level to the 3' coding sequences of several reported proton (H+) ATPases
(Harper et aI., 1989; Pardot and Serrano, 1989; Ewing et al., 1989; Harper et al.,
1990 and 'Wada et aI., 1992) The predicted protein sequence of this DNA region
(approximately 80 residues) showed even higher homologies with known H+-
t4r
ATPases (Figure 6.8). The sequence 3' distal to this region of high homology appears
to be non-coding. It contains an in-frame stop codon and the protein homology
between the sequence and the reported H+-ATPases abruptly stops at this point .
The BLAST sea¡ch did not locate any homology between the DNA sequence of BMR6
and any well characterised genes. However, two regions of the putative protein
encoded by this cDNA show homology (55-657o positives) to a scenescence-related
protein from radish, din I and a stress-related protein from E. coli,pspE, both of
which are induced under a number of environmental conditions (Azumi and \Vatarrabe,
I99l; Brissette et a1.,1991).
FIGURE 6.6
Restriction mapping and sequencing strategies for BMR6 and BMR78
The phage inserts were cloned into the Eco Rl site of pTZlSU. Restriction maps were
prepared by endonuclease-digestion analysis of pBMR6 and pBMR78. The inserts of
the original clones and sub-clones containing various restriction fragments were
sequenced as indicated. The arrows indicate the direction and extent of each
sequencing reaction.
E =Eco Rl, R= Rsa 1, P= Pst 1, S = Sac I andX = Xba 1.
Figure 6.6
pBMRó
pBMR78
3s0
l00bp
ERr20
ER
+
165
+
150R
->
170
R
155
XSSE110
ESX
FTGURE 6.7
The nucleic acid and predicted protein sequences of BMR6 and BMR78
The predicted protein sequence is shown under the open reading frame for each clone
and the stop codon in each sequence is marked with an asterisk (*). Putative poly-
adenylation signal sequences (AATAA) in the untranslated 3' regions of each clone are
underlined as is a putative C-terminal phosphorylation site (see Ha¡per et al., 1990) in
the predicted protein sequence of BMR78.
BMR6 (657 bp)
CCT CTG GAG AGC ACC ACA TÀC GGA ÀCC GGA ÀGC TCT TTC CTC CCC GGTPLESTTYGTGSSFLPG
GCT C.A.A TCC TCC ÀTA GAT GGG CTC CCT GCG AAG ACA GTG AGC ACG GCGAOSSIDGLPAKTVSTA
GTG GAC AGC GTG GAC GCG GAG GCG GCG TGC GCA CTG CTG GCC TTG GAGVDSVDAEÀACALI,A],E
CAG TAC GGG TAC GTG GAC GTG CGG ATG TGG GAG GAC TTC GAÀ AÀG GGTOYGYVDVRMWEDFEKG
CAC GTT GCC GGT GCA CGC AAC GTG CCC TAC TAC CTG TCC GTC AGC GCCHVAGARNVPYYLSVSA
CCA TGG CAA GGA GCG CAA CCC GGA CTT CGG TCG ÀTC AGG TTG CGG CGCPWOGAOPGI,RSTRLRR
TCC ACT CCA AGG AGG ACC GGT TCC TTG TTG GTT GTC GCT CTG GGG TCCSTPRRTGSLLVVALGS
GGT CCA GGC TCG CCAGPGSP
CCG CAG ACC TTG TCG CCG CTG GAT TCG CA.A, ACGPQTLSPLDSQT
TGA AGA ACC TCG AÀG GCG GCT ÀCC TCT CCT TGC TCA AGA GCG CAA GTT*
Àr\r\
GGALLtL
ACGGTÀ
A¡,GAAGCTTGCTssI
CAGTTTATG(Ju(J
ACG
À^a('r\UAAGTGTLrb¡\
uAa
CTGAAGAGTACT
GATTTTAAGIUå
LrUlJ
ccAuauAAGAÀGGÃÀ
IUåtåuCAÀGGAAÀÀ
ATTTATGTCAÀAAÀÄ
uårGTCCTTCAAA
TCTTCTTÀGtrtrU
TTGGTCU.ILJ
ÀTT
(JA(J
ATC¡llJtr
ATG
TTTCTTTGCTAT
ccA cccUA ] AUf
A.AG A¡,GATT GAAÀÀ'l ÀÀT
BMR78 (570 bp)
A.A.ê, GGT GAG AGG GÀG GCA CAGKGEREAO
GGC CTT CAA GCA CCT GAG CCGGLQAPEP
AGC AGC TAC CGT GAG CTT TCTSSYRELS
GÀG TCC GTG GTG AAG CTC AAGESVVKLK
A.AC TAC ÀCC GTG TGA GGA ATTNYTV*
TGG GCT ACC GCA CAG AGA ACA CTT CATV{ïATAORTLH
GCT TCC CAC ACA CTC TTC AAC GAC AÂGASHTLFNDK
GAG ATC GCT GAG CAA GCC A.AG AGA AGAEIAEQAKRR
GGC CTT GAC ATC GAC ACC ATC AÄC CÀÄGLDIDTTNO
TGG GCA TTG GAC CAA ACA ACA TGA CAT
GCT GAG ATT GCA AGG TTG AGG GAG CTC AÀC A,CA CTC AAG GGG CAC GTTAEIARLRELNTLKGHV
CCG TAT TGGGTG CAT GAAAAG TGC TACATG TCT AGTTTl ACC CTTAÀA_TAÀICA
¿ÀLrU
TGAC.AÀGGÀfuu\/r\f
GAGA.AG
ÀTAGTTccrCAA
ÀTTAÀATATATAAACÀTT
Lrtr¡\CATGGT\-UtrTAGTTC
åau
lJtrL
TTCTGTTCGTGC
(J/t(J
f tJft
GGTTTGGTGCAA
crcTCTTGTGAGATCÀAÀ
UtrUGACLUUUJIL-
TTGAAÀ
TTTCAGTCT\fl-Lr
\-!J(J
AÄA
AA,G
cccAGA-! \JtJ
AÀGAAA
AIå
CTAGÀ,G
aurCAAA.A.À
TTTGTTGTTAATGAC
ccGuaåTTAULrLt
GAG
FIGURE 6.8
Alignments and summary of homologies with the predicted proteins of the BMR78
(Figure 6.8a) and BMR6 (Figure 6.8b) cDNA sequences.
The translations, alignments and retrieval of homologous sequences were carried out
using the BLAST (Altschul et aI., 1990) program. The predicted protein sequence of
BMR78 shows 80-90Vo homology with the C-terminal region of all known plant
ATPases. BMR6 has two regions which have homologies (55-65Vo) to genes from
E.coli and radish.
Figure 6.8a
BMR78STDPHAlSTDP}IA2NPPMA4ATHHATPAl-ATH}IATPA2ATHHATPA3RICOSAlTOMLHAl-TOMHA2
187687I877868868870876876624
KGEREAQWATAQRTLHGLQAPEPASHTLFNDKSSYRELSEIAEQAKRRÀEIARLRELNTLKG}IVESWKLKGLDIDTINQNYTV*E+**L***H*********V*+-TK--**++S+++N**+++:k*a*********Q***.r*****************+**e*+****E*******L*********P**-A*N-***+*+*****************a******.r:t****************+**e*+****E*******L*********P**-ATN-*:k*-rr*-u*****************+******..,r*****************.r_**e*+***I********Q*********PK*-AVN-+*P+*G**********************:k*.r-*********A*********AGH+****E***rç***L*********PK*-A\iîü-+*p+KG*Jr**********************+*****************+*pSH_***f E* * * * * * *H* * * * * * * * *NT*Tr* +W++- -RGG* * * * * * * *N* * * * * * * * * * * * * *+ * * * * * * * * * * * * * * * * *.r * AGH_ * * **E***L+**H*H*******P*+-i\I(--P*P+*+G*S**+++**+*************.u********************+*+****E***L***H*********V*+*-K--+*++T+++N**+++**.u*************+*****************+**e*+****EQ**L***H*********V*+-TrK--**++A+++N**.raa**-rr*********G***+*****************.r-**e*+***
83956952952949942944956956704
Figure 6.8b
BMR6 45Din 1, 82Psp E 16
LAIEQYGYVDVRMVVEDFEKGHVAGARNVPYYLSV 7 B**QAG*K++* **TP++*SI* *P+R*I** **MYR* 1I7PVFAAEH++ * * *+p*Q+++E* *Q* * I *+PLKEVK 51
KEDRFLVGCRSGVRS RLATADLVAÄ,GFANVKNLEGGYL S LLK SASYQAÀSHRS I*H*EI++* *E* *E* *L+*+T+ *+T* * *TG*T++A* * *+*N*T\T(*Y*N+ *R+ *GQ*KEI *SEM*+TH+ *+ *AG*LK
BIIR6Din 1Psp E
991-376I
1_53186106
t45
6.6 Discussion
These results provide evidence that the cDNA clones BMR6 and BMR78 detect root-
specific, low-copy plant genes which are specifically upþgulated during the ÀA
establishment of the mycorrhizal symbiosis with G. intraradices and not by general
biotic or abiotic conditions.
The doubled haploid and addition line mapping experiments have placed the genes
represented by BMR6 and BMR78 on barley chromosomes 6H andZH respectively.
BMR78 and AWBM21 (another RFLP marker which was isolated from the M+ barley
root cDNA library) map to a region of 2H near the Ha2 gene (4. Karakousis et aI.,
unpublished data) which confers resistance to a cereal cyst nematode (CCN). At a
purely applied level, these two markers could prove useful in experiments aimed at
isolating the resistance gene. At the very least these cDNA clones provide further
coverage of the barley genetic map and could represent useful markers for CCN
resistance.
There seems little doubt that BMR78 represents the partial sequence of a H+-ATPase
and is the first such sequence isolated from H. vulgare. This clone detects root-
specific mRNA transcripts which increase in abundance during mycorrhizal
development. These results support the data provided by enzymatic and
immunolabelling experiments which show that membrane-bound ATPase activity is
increased in the host plant-arbuscule interface (Man et a1.,1982; Gianninazzi-Pearson
et al.,1991) and add further weight to the argument that this intertäce is involved in
the active transport of solutes between the symbiotic partners (see Smith and Smith,
1990) and as such, may play a specific role in the development and/or maintenance of
the symbiosis.
Analyses of the complete-gene sequences for proton ATPases from Nicotiana
plumbaginoþlia, Lycopersicon esculentum andArabidopsis thaliana, show them to
be multigene families which are highly conserved both within and between species
t46
(Boutry et aI., 1989; Ewing et a1.,1990; Harper et a1.,1990). A.lçhatiana, the best 't
/\
characterised of the H+-ATPase families, contains at least 10 isoforms of the gene,
some of which are tissue specific (Harper et al., 1994; Houlné and Boutry, 1994).
The mapping experiments provide some evidence that the gene corresponding to
BMR78 may be a member of a multigenic family in which case the specificity of
expression of this gene (and any other isoform) needs to be more completely assessed.
Reprobing 20000 clones from the M+ library (as for the differential screening
described in Chapter 5) with radiolabelled probe made from BMR78, identified 12
strongly hybridising phage clones and 7 weakly hybridising phage clones. PCR
amplification and subsequent Xba I restriction analysis of six phage from each group
indicated that the twelve strongly hybridising phage were other clones of or very
similar to BMR78 whilst the seven weakly hybridising phage were a different
sequence (results not shown). These results remain inconclusive in the absence of a
more detailed analysis of these clones but nevertheless indicate that at least two types
of proton ATPases may be expressed in the roots of mycorrhizal barley roots. This
raises the question of whether BMR78 represents an isoform of the gene which is
specifically upregulated during the development of mycorrhizas. There is also a
possibility that the constitutive expression observed in the original Northern
hybridisations may reflect cross-hybridisation between BMR78 and different root H+-
ATPase genes, all of which produce mRNA transcripts of approximately the same
length. These are both important and interesting questions which warrant further
detailed and more specific attention.
The characterisation of BMR6 yielded far less information. It has two virtually
continuous regions of homology with genes isolated from radish artd E. coli which
can be induced by a range of stress-like conditions. These two regions from BMR6
are separatedby 2I amino acid residues. Similarly, the regions of homology identified
in the din I and psp E genes are only separated by 20 and l0 amino acid residues,
respectively (see Figure 6.8). As water-stress conditions and pathogen-infection did
r47
not greatly induce accumulation of the transcripts detected by this cDNA sequence (see
Figure 6.1), the gene BMR6 represents is not thought to be a general stress-induced or
defence-related gene. As such, these two regions of homology may indicate a coÍìmon
secondary or tertiary structural feature of the three gene products rather than a conìmon
functional region.
The abundance of gene transcripts detected by BMR6 does not appear to greatly
increase in response to either pathogen infection or water-stress. It is possible, indeed
likely, that BM6 detects the RNA transcripts of a gene which is induced purely in
response to the mycorrhiza colonisation process and that it is not a symbiosis-related
gene actively involved in the developmental process itself. Further and more extensive
expression analyses using RNA extracted from various barley tissues grown under a
range of environmental conditions may provide better evidence of the biological
function of the gene product. The biological significance of the gene remains unclear.
148
Chanter 7
General Discussion
7.1 Model plant-fungus systems for
investigating mycorrhizal symbioses.
Mycorrhizal symbioses are the culmination of a complex sequence of events which
are genetically determined by both the plant and fungal partners. Elucidating the
molecular genetic events which result in the formation of a compatible (ie.
mutualistic) mycorrhizal association would enrich our understanding of these
symbioses.
VA mycorrhizal fungi are obligate symbionts and efforts to culture them in vitro
have been only partially succesful. At present these interactions can only be studied
using model plant/fungus systems grown in some type of pot culture. The VA
mycorrhizal system developed during the course of this work used Hordeum vulgare
(barley) cv. Galleon as the host plant and Glomus intraradices as the fungal partner.
Molecular analyses are extremely sensitive in comparison with many biochemical
and enzymatic methods but still have finite limits of detection. This work (and that of
Baon et al., 1994) showed that considerable phenotypic variation exists between
local South Australian barley cultivars with respect to the rate and extent of root
colonisation by Glomales species. This is in agreement with published data for
several other plant species (see Smith, Robson and Abbott, 1992) and was an
important consideration because the success of the approach described in this work
was dependant upon being able to reliably produce substantial quantities of heavily-
colonised mycorrhizal root tissue. This was difficult to achieve over the whole barley
root-system, but root tissues 'enriched' for the early myconbizal colonisation stages
t49were obtainable by manipulating the levels of inoculum in the pot system and
studying the time-course and distribution of colonisation. Some recent work by
Rosewarne et aI. (pers. comm.) with tomato seedlings indicates that a 'nurse plant'
pot culture system may be an easier and more effective means of producing large
amounts of early-stage colonised VA mycorrhizas.
Barley has a fast growing root system which produces sufficient quantities of RNA
for the types of analyses undertaken in this work. The extent of colonisation using
this model system was highly reproducible within any given batch of inoculum.
However, variation did exist between batches of inoculum. Some produced poor
levels of colonisation in 14 day old barley seedlings and were immediately rejected
to be used only for inoculum production. The wastage of several inoculum batches
was not a problem because a large amount of inoculum was available (courtesy of
Dr. S. Smith and S. Dixon) but does need to be considered if the resources available
are more limited.
7.2 Towards identifying Myc mutants.
The importance of identifying and characterising plant phenotypes defective in
mycorrhiza formation or function cannot be overstated (see Koornneef, 1991 and
Kyla, 1994). They are the key to elucidating many of the developmental events, at
both the molecular and physiological levels, that lead to a successful mutualistic
interaction. Considerable research effort has gone into developing strategies (such as
gene-tagging) which simplify either the generation of mutations or more importantly,
the subsequent isolation of mutant genes from plants. A major problem still remains
(and it is emphasised by this work; obtaining sufficient resources (chiefly human) to
produce and analyse the large number of plants required to have a high probability of
success.
Screening for barley-mycorrhiza mutants was unsuccessful largely because only
1800 plants from the two mutagenised 'Galleon' populations were screened.
Nutritional screening for non-functional mycorrhizal status was attempted but was
150
unsuccessful. A large percentage of plants from all the treatment groups (ie. with or
without added phosphorus and/or mycorrhizal inoculum) showed P deficiency-like
symptoms which remained till the time of harvest (3-4 weeks). It is presumed that
the fast growth rate of barley roots (and thus requirement for P) outstripped the extra
P made available by having mycorrhizal roots. As a consequence, the roots of each
seedling had to be individually assessed for extent and type of mycorrhizal
colonisation using a microscope.
Many of the mycorrhizal and phosphate-amended plants grown in pots containing
1kg of soil medium (see Chapter 1) did recover their normal colour after initially
displaying P deficiency-tike chlorosis. However,large-scale nutritional screening for
myc- mutarús under these conditions would require several tonnes of soil medium for
each experiment as well as large areas of growth room space for an extended period
of time (4-6 weeks). The resources required to set up and manage screening
experiments on this scale are not readily available.
Individual screening of a large number of plants for mycorrhizal mutant phenotypes
is obviously expensive in terms of time and resources and is unlikely to be directly
funded. Therefore, it is difficult to see much progress being made in this area of
mycorrhizal research other than by serendipidy (much like the nod- /myc- mutations
described by Duc et a1.,1989) unless a quick and simple assay (possibly nutritional)
can be devised which allows the screening of many thousands of plants at once. In
hindsight, barley may not be the best plant for this type of approach. A slower-
growing host plant, suitable for genetic manipulation and analysis, coupled with an
aggressive Glomales species may provide a more amenable model system for the
detection of plant mutants defective in mycorrhiza development and./or function.
7.3 Molecular approaches to
cloning functional mycorrhiza genes
These experiments have shown that it is practicable to use molecular approaches to
investigate the mycorrhizal symbiosis. Over 1000 putative clones were detected after
151
the initial diffrential screening, this was subsequently reduced to I2l after a second
round of screening. DNA:RNA hybridisation analysis of 46 of these clones identified
four cDNA clones that showed changes in expression patterns which correlated with
colonisation of roots by Glomus intraradices. Two of these mRNA species are
down-regulated (decreased accumulation) and two are up-regulated (increased
accumulation) during mycorrhizal development.
The sequence analysis of the two upregulated clones indicates that the putative
protein of BMR6 may be a stress-related protein and that BMR78 is highly
homologous to a multigenic family of proton ATPases isolated from a range of plant
species. These results are not surprising as there is considerable physiological and
biochemical evidence which shows plant pathogenic reactions and increased ATPase
activity occurs in response to mycorrhizal colonisation. S/hether these clones are
essential to the formation and maintenance of a fuctional mycorrhiza is not known
but they may prove useful in monitoring and defrning certain stages of colonisation.
A clearer picture of the temporal and spatial expression of the two genes that BMR6
and BMRT represent may emerge if the clones are used for in situ hybridisation
experiments. Similarly, purification of the proteins (in an Escherechia coli
expression system) and preparation of antibodies should allow precise localisation of
the proteins within root sections. It may also be possible to use transgenic
approaches to'knockout'or overexpress genes in barley and assess the effects on
planlVA mycorrhiza phenotype.
More extensive cloning and screening should yield additional sequences of interest.
A targeted approach, using cloned sequences from other organisms as heterologous
probes, should also provide further basic information on the interaction as well as
producing additional probes for monitoring developmental stages.
7.4 Concluding statements
This is the flust time an untargeted cloning approach has been used to investigate the
VA mycorrhizal symbiosis at the molecular level. Whilst the mutagenesis
r52experiment failed to identify any obviously useful phenotypes, it nevertheless
demonstrated that screening for myc- mutations is feasable as it appears that a large
number of genes may have an effect on the development of myconhizas. Principally,
this work has provided good molecular-genetic evidence that plant-root gene
expression is altered during the ontogeny of these specialised structures. The
barleytGlomus model system which was developed could provide a basis for future
studies of this interaction using other plant-VA myconhizal fungus combinations.
Mycorrhizas appear to be important in the nutritional dynamics of nearly all
ecosystems. Studying the symbiosis in the intimate detail afforded by molecular
analyses may eventually allow manipulation of VA mycorrhizal fungi at the
agricultural field level. More importantly though, it may give us considerable insight
into the evolution of plant-micobial interactions and the mechanisms underlying
cellular communication. This work represents a small but important first step.
153
References
Abbott L.K., Robson A.D. and de Boer G. (1984). The effect of phosphorus on the
formation of hyphae in the soil by the vesicular-mycorrhizal fungus Glomus
fasiculatum. New Phytol. 97 : 437 446
Abu-Abied M. and Holland D. (1994). The gene c-inol from Citrus paradisi is
highly homologous to turl and inol from yeast and Spirodel¿ encoding for myo-
inositol phosphate synthase. Plant Physiol. 106: 1689
Alexander T., Toth R., Meier R. and'Weber H.C. (1989). Dynamics of arbuscule
development and degeneration in onion, bean and tomato with reference to vesicular-
a¡buscular mycorrhizae with grasses. Can. J. Bot. 67:2505-2513
Allen M.F. (1982). Influence of vesicula¡-arbuscular mycorrhizae on water movement
through Bouteloua gracilis (fßK) Lag x Steud. New Phytol. 91: 191-196
Allen M.F., Moore T.S. and Christensen M. (1980). Phytohoûnone changes in
Bouteloun gracilis infected by vesicular-arbuscular mycorrhizae. I. C¡okinin increases
in the host plant. Can. J. Bot. 58: 371-374
Allen M.F., Smith W.K., Moore T; and Christensen M. (1981). Comparative water
relations and photosynthesis of mycorrhizal and non-mycorrhizal Bouteloua gracilis
(IßK) Lag x Steud. New Phytol. 88: 683-693
Altschul S.F., Gish'W., Miller'W., Myers E.V/. and Lipman D.J. (1990). Basic local
alignment search tool. J. Mol. Biol. 215:4034L0
Ames R.N., Porter L., St. John T.V. and Reid C.P.P. (1984). Nitrogen sources and
'A' values for vesicular-arbuscular and non-mycorrhizal sorghum grown at three rates
of 15-N ammonium sulphate. New Phytol.97:269-276
t54
Ames R.N., Reid C.P.P., Porter L. and Camba¡della C. (1983). Hyphal uptake and
transport of nitrogen from two l5-N-labelled sources by Glomus mosseae, a vesicula¡-
arbuscula¡ fungus. New Phytol. 95: 381-396
Amijee F., Tinker P.B. and Stribley D.P. (1989). The development of
endomychomzalroot systems: VII. A detailed study of effects of soil phosphorous on
colonisation. New Phytol. lll: 435446
Anderson A.J. (1988). Myconhizae-host specificity and recognition. Phytopath. 78:
375-378
Arnheim N. and Erlich H. (1992). Polymerase chain reaction strategy. Ann. Rev
Biochem. 52:131-155
Azcon R. and Ocampo J.A. (1981). Factors affecting the vesicular-arbuscular
infection and mycorrhizal dependency of thirteen wheat cultiva¡s. New Phytol. 87:
677-685
Azcon R., Marin A.D. and Barea J.M. (1978). Comparative role of phosphate in soil
or inside the host on the formation and effects of endomycorrhiza. Plant Soil 49: 561-
567
Azumi Y. and Watanabe A. (1991). Evidence for a scenescence-associated gene
induced by darkness. Plant Physiol. 95:577-583
Baker B., Schell J.,Lotz H. and Federoff N. (1986). Transposition of the maize
controlling element "Activator" in tobacco. Proc. Natl. Acad. Sci. (USA) 83: 4844-
4848
Balcells 8., Swinburne J., Coupland G. (1991). Transposons as tools for the
isolation of plant genes. Trends in Biotech. 9:31-37
155
Baon J.B., Smith S.E. and Alston A.M and'Wheeler R.D. (1992). Phosphorus
effîciency of three cereals as related to indigenous mycorrhizal infection. Aust. J.
Agric. Res.43: 479491
Baon J.B., Smith S.E. and Alston A.M. (1993). Mycorrhizal responses of barley
cultivars differing in P eff,rciency. Plant and Soil 157:97-lO5
Bartchi H., Gianinazzi-Pearcon V. and Vegh I. (1981). Vesicular-arbuscular
mycorrhiza formation and root rot disease (Phytophthora cinnamoni) development in
C haemae cy p aris lnw s oniana. Phytopath. Z. IO2:. 213 -218
Bauer \M.D. (1981). Infection of legumes by rhizobia. Ann. Rev. Plant Physiol. 32:
40749
Bécard G. and Piché Y. (1989). New aspects on the acquisition of biotrophic status by
a vesicular-arbuscular mycorrhizal fungus Gigasporamargarita. New Phytol.lL2:77-
83
Bécard G. and Piché Y. (1990). Physiological factors determining vesicular-
a¡buscular mycorrhizal formation in host and nonhost Ri T-DNA transformed roots.
Can. J. Bot. 68: 1260-12&
Bécard G. Douds D.D. and Pfeffer P.E. (1992). Extensive in vitro hyphal growth of
vesicula¡-arbuscular mycorrhizal fungi in the presence of CO2 and flavanols. Appl.
Envi¡. Microbiol. 58: 821-825
Beckman J.S. and Soller M. (1990). Toward a unified approach to genetic mapping of
eukaryotes based on sequence tagged microsatellite sites. Bio/Technology 8: 930-932
Bel Rhid R., Chabot S., Piché Y. and Chênevert R. (1993). Isolation and
identification of flavanoids from Ri T-DNA-transformed roots (Dazcus carota) and
their significance in vesicular-a¡buscula¡ mycorrhiza. Phytochem.33: L369-1371
156
Berta G., Sgorbati S., Soler V., Fusconi 4., Trotta 4., Citterio 4., Bottone M.G.,
Sparvoli E. and Scannerini S. (1990). Variations in chromatin structure in host nuclei
of a vesicular arbuscular mycorrhiza. New Phytol. ll4: 199-205
Bevege D.I. and Bowen G.D. (1975). Endogone strain and host plant differences in
development of of vesicula¡-arbuscular mycorrhizas. In: Endomycorrhizas; Sanders
F.E., Mosse B. and Tinker P.B. (eds.), Academic Press, London and New York. pp
77-86
Bhuvaneswari T.V. and Solheim B. (1985). Root hair deformations in the white
clover I Rhizobium trtfolü symbiosis. Physiol. Plant. 63:25-34
Bhuvaneswa¡i T.V., Turgeon G.B. and Bauer W.D. (1981). Early events in the
infection of soybeans (Glycine max L. Mer). Plant Physiol. 66: t027-1031
Bieleski R.L. and Ferguson I.B. (1933). Physiology and metabolism of phosphate
and its compounds. In: Encyclopaedia of Plant Physiology, Vol. 154: Inorganic Plant
Nutrition; Lauchli A. and Bielesþ R.L. (eds.), Springer Verlag, Berlin. pp 422449
Bolton G.'W., Nester E.\V. and Gordan M.P. (1986). Plant phenolic compounds
induce the expression of the Agrobacterium tumefaciens loci needed for vi¡ulence.
Science 232:983-985
Bonfante-Fasolo P. (1987). Vesicular-arbuscular mycorrhizae: Fungus-plant
interactions at the cellular level. Symbiosis 3: 249-268
Bonfante-Fasolo P. (1988). The role of the cell wall as a signal in mycorrhizal
associations. In: Cell to Cell Signals in Plant, Animal and Microbial Symbioses; S.
Scannerini, D. Smith, P. Bonfante-Fasolo, V. Gianinazzi-Pearson (eds.), Springer-
Verlag, Berlin, Heidelberg. pp 6a-78
t57
Bonfante-Fasolo P. (1988). The role of the cell wall as a signal in mycorrhizal
associations. In: Cell to Cell Signals in Plant, Animal and Microbial Symbiosis;
Springer-Verlag Berlin Heidelber g. pp 219 -235
Bonfante-Fasolo P. and Grippiolo R. (1982). Ultrastructural and cytological changes
in the wall of a vesicular-arbuscular mycorrhizal fungus during symbiosis. Can. J.
Bot.60: 2303-2312
Bonfante-Fasolo P., Dexheimer J., Gianinazzi-Pearson V. and Scannerini S. (1981).
Cytochemical modifications in the host-fungus interface during intracellular
interactions in vesicula¡-arbuscular mycorrhizae. Plant Sci. Lett. 22: 13-21
Bonfante-Fasolo P., Vian 8., Perotto S., Faccio A. and Knox J.P. (1990). Cellulose
and pectin localisation in roots of mycorrhizal Allium porumi labelling between the
host cell wall and interfacial material. Planta 180:537-547
Bonfante-Fasoslo P. and Spanu P. (1992). Pathogenic and endomycorrhizal
associations. In: Methods in Microbiology Volume 24. Academic Press Ltd. London.
pp 14l-169
Bonnet H.T., Djurberg I., Fajardo M. and Glimelius K. (1993). A mutation causing
variegation and abnormal development in tobacco is associated with an altered
mitochondrial DNA. Plant J. 3:519-525
Borojevic K., Gottschallk W. and Micke A. (1977). Factors influencing the mutant
spectrum and quality of mutants. In: Manual on Mutation Breeding, IAEA, Vienna.
pp. 146-150
Boutry M., Michelet B. and Goffeau A. (1989). Molecula¡ cloning of the family of
plant genes encoding a protein homologous to plasma membrane H+-translocating
ATPases. Biochem. Biophys. Res. Commun. 162:567-574
158
Bowles D.J. (1990). Defense-related proteins in higher plants. Ann.Rev. Biochem.
59 873-907
Bowman J.L., Smyth D.R. and Meyerowitz E.M. (1989). Genes directing flower
development ín Arabidopsis. Plant Cell I : 37-52
Boyetchko S.M. and Tewari J.P. (1990). Root colonisation of different hosts by the
vesicula¡-arbuscular mycorrhizal fungus Glomus dimorphicum. Plant Soil 129: 131-
t36
Bradbury S.M., Peterson R.L. and Bowley S.R. (1991). Interactions between th¡ee
alfalfa nodulation genotypes and nvo Glomus species. New Phytol. I 19: ll5-l20
Briggs R.W. and Constantin M.J. (1977). Radiation types and radiation sources. In:
Manual on Mutation Breeding,IAEA, Vienna, pp.7-21
Brissette J.L., Weiner L., Ripmaster T.L. and Model P. (1991). Characterisation and
sequence of the Escherichia coli süess-inducedpsp operon. J. Mol. BtoL.220;35-48
Brown T.A. (1991). Molecular Biology: Labfax. Bios Scientific Publishers,
Blackwell Scientific Publications, Oxford.
Brundrett M.C., Piché Y. and Peterson R.L. (1985). A developmental study of the
early stages in vesicula¡-arbuscular mycorrhizal formation. Can, J. Bot. 63: 184-194
Bryla D.R. and Koide R.T. (1990). Role of mycorrhizal infection in the gowth and
reproduction of wild vs.cultivated plants tr. Eight wild accessions and two cultiva¡s of
Lycopersicon esculentum Mill. Oecologia 84: 82-92
Cabrera y Poch H.L., Peto C.A. and Chory J. (1993). A mutation on the Aribidopsis
DET3 gene uncouples photoregulated leaf development from gene expression and
chloroplast biogenesis. Plant J. 4: 67 l-682
159
Caetano-Anollés G. and Gresshoff P.M. (1992). Plant genetic suppression of the non-
nodulating phenotype of Rhizobium meliloti host range nodH mutants: gene-for-gene
interaction in the alfafia-Rhizobium symbiosis? Theor. App. Genet.84:624-632
Carling D.R., Riehle W.G., Brown M.F. and Johnson D.R. (1978). Effects of a
vesicula¡-a¡buscular mycorrhizal fungus on niEate reductase and nitrogenase activities
in nodulating and non-nodulating legumes. Phytopath. 68: 1590-1596
Carr G.R., Hinkley M.4., Le Tacon F., Hepper C.M., Jones M.G.K. and Thomas
E. (1985). Improved hyphal growth of two species of vesicular-arbuscular
mycorrrhizal fungi in the presence of suspension cultured plant cells. New Phytol.
lO1.:417426
Cervone F., Castoria R., Spanu P. and Bonfante-Fasolo P. (1988). Pectinolytic
activity in some ericoid mycorrhizal fungi. Trans. Br. Mycol. Soc. 9l: 537-539
Chen J. and Varner J.E. (1985). Isolation and cha¡acterisation of cDNA clones for
carrot extensin and a proline-rich 33-kDa protein. Proc. Natl. Acad. Sci. (USA) 82:
4399403
Christou P., Ford T.L. and Kofron M. (1992). The development of a variety-
independant gene-transfer method for rice. Trends in Biotech. lO:239-246
Collinge D.B. and Slusarenko A.J. (1987). Plant gene expression in response to
pathogens. Plant Mol. Biol. 9: 389410
Cooke R.J. (1988). The use of electrophoresis for the distinctness testing of va¡ieties
of autogamous species. Plant Va¡. Seeds 2:3-13
Cooper K. (1935). Physiology of VA mycorrhizal associations. In: VA Mycorrhizas;
C.L. Powell and D.J. Bagyaraj (eds.), CRC Press, Boca Raton, Florida. pp 165-183
160
Cordignola 4., Verotta L., Spanu P., Maffei M., Scannerini S. and Bonfante-Fasolo
P. (1989). Cell wall bound-phenols in roots of vesicular-arbuscular mycorrhizal
plants. New Phytol. Il2 221-228
Cove D.J. (1993). Mutant analysis, a key tool for the study of metabolism and
development. Plant J. 3: 303-308
Cox G. and Sanders F.E. (1974). Ultrastructure of the host-fungus interface in
vesicula¡-arbuscular mycorrhiza. New Phytol. 7 3: 901 -9 12
Cox G. and Tinker P.B. (1976). Translocation and transfer of nutrients in vesicular-
arbuscular mycorrhizas. I. The a¡buscule and phosphorus transfer: a quantitative
ultrastn¡ctural study. New Phytol. 7 7 : 37 | -37 I
Cress W.4., Throneberry G.O. and Lindsey D.L. (1979). Kinetics of phosphate
absorption by mycorrhizal and non-mycorrhizal tomato roots. Plant Physiol.64:484-
487
Crush J.R. and Caradus J.R (1980). Effect of mycorrhizas on growth of some white
clovers. N.Z. J. Agric. Res. 23: 233-237
Daniels B.A. and Graham S.O. (1976). Effects of nutrition and soil extracts on
germinationof Glomus mosseae spores. Mycologia 68: 108-l16
Darlington C.D. and Mcleish J. (1951). Action of maleic hydrazide on the cell.
Nature (London) 167 : 4O74O8
Darnell J.E. (1982). Variety in th level of gene control in eukaryotic cells. Nature
(London) 285:365-371
Davidson 4.D., Manners J.M., Simpson R.S. and Scott K.J. (1987). cDNA cloning
of mRNAs induced in resistant barley during infection by Erysiphe graminis f..sp.
Hordei. Plant Mol. Biol. 8: 77-85
161
Davies E. and Schuster A. (1981). Intercellular communication in plants: Evidence for
a rapidly generated, bidirectionally transmitted wound signal. Proc. N. Acad. Sci.
(USA) 78:2422-2426
De Pater S., Hensgens L.A.M. and Schilperoort R.A. (1990). Structure and
expression of a light-inducible shoot-specific rice gene. Plant Mol. Biol. 15:-3994O6
Dénaré J., Debellé F. and Rosenberg C. (1990). Signalling and host range variation in
nodulation. Ann. Rev. Microbiol.46: 497-531
Dexheimer J., Gianinazzi S. and Gianinazzi-Pearson V. (1979). Ultrastructural
cytochemistry of the host-fungus interface in the endomycorrhizal association Glomus
mo s s e ae/Allium c e p a. Z. Pflatuenphysiol. 92: 19 I -206
Dixon R.A. and Lamb C.J. (1990). Molecular communication in interactions between
plants and microbial partners. Ann. Rev. Plant Physiol. Plant Mol. Biol.4l: 339-367
Dixon R.4., Ha¡rison M.J. and La¡nb C.J. (1994). Early events in the activation of
plant defence responses. Ann. Rev. Phytopath. 32: 479-50L
Djordjevic M.4., Gabriel, D.rW. and Rolfe, B.G. (1987). Rhizobium - the refined
parasite of legumes. Ann. Rev. Phytopath. 25: 145-168
Duc G., Trouvelet 4., Gianinazzi-Pearson V. and Gianinazzi S. (1989). First report
of non-mycorrhizal plant mutants (Myc-) obtained in Pea (Pisum sativa L.) and
Fababean (Viciafaba L.). Plant Science 60 215-222
Dumas E., Gianinazzi-Pearcon V. and Gianinazzi S. (1989). Production of new
soluble proteins during VA mycorrhiza formation.Agric.Eco. Env. 29: lll-l14
Dumas-Gaudot E. Furlan V. Grenier J. and Asselin A. (1992). New acid chitinase
isoforms induced in tobacco roots by vesicular-arbuscular mycorrhizal fungi.
Mycorrhiza l; 133-L36
t62
Elias K.S. and Safir G.R. (1987). Hyphal elongation of Glomus fasciculatus in
response to root exudates. Appl.Env. Microbiol. 53: 1928-1933
Ellingboe A.H. (1981). Changing concepts in host-pathogen genetics. Ann. Rev.
Phytopath. 19: L25-143
Ellis J.G., Lawrence G.J., Peacock V/.J. and Pryor A.J. (1988). Approaches to
cloning plant genes conferring resistance to fungal pathogens. Ann. Rev. Phytopath.
26 245-263
Ellis R.J. (1990). Molecular chaperones: the plant connection. Science 25O:954-959
Estaun V., Calvert C. and Hayman D.S. (1987). Influence of plant genotype on
mycorrhizal infection: Response of th¡ee pea cultivars. Plant and Soil 103:295-298
Evans C.T., Sumegi 8., Srere P.4., Sherry A.D. and Malloy C.R. (1993). (13C)
propionate oxidation in wild-type and citrate synthase mutant Escherichia coli :
evidence for multiple pathways of propionate utilisation. Biochem. J.29I: 927 -932
Ewing N.N., Wimmers L.E., Meyer D.J., Chetelat R,T. and Bennett A.B. (1990).
Molecular cloning of tomato plasma membrane H+-ATPase. Plant Physiol. 94:. 1874-
1881
Federoff N.V., Furtek D.B. and Nelson O.E. Jn. (1984). Cloning of the bronze
locus in maize by a simple and generalisable procedure using the transposable element
Activator (Ac). Proc. Natl. Acad. Sci. (USA) 8l:3825-3829
Feldmann K.A. (1991). T-DNA insertion mutagenesis in Arabidopsis : mutational
spectrum. Plant J. t:71-82
Flor H.H. (1971). Current status of the gene-for-gene concept. Ann. Rev. Plant Path.
9:275-296
163
Folch J.L., Antaramian 4., RodriguezL., Brava A. and Brunner G. (1989). Isolation
and characterisation of asaccharomyces cerevisiae mutant with impaired glutamate
synthase activity. J. Bacteriol. 171: 6776-6781
Franken P. and Gnadinger F. (1994). Analysis of parsley arbuscular mycorrhiza:
Infection develpment and mRNA levels of defense-related genes. Mol. Plant-Mic.
Inter. T: 612-620
Friese C.F. and Allen M.F. (1991). The spread of VA mycorrhizal fungal hyphae in
the soil: inoculum types and extemal hyphal architecture. Mycologia 83: 409-418
Garbaye J. (1994). Helper bacteria: a new dimension to the mycorrhizal symbiosis.
New Phyt. t28: L97-210
Ga¡cia-Romero I., Garcia-Ga¡rido J.M. and Ocampo J.A. (1991). Pectinase activity in
vesicular-a¡buscula¡ mycorrhiza during colonisation of lettuce. Symbiosis 12: 189-198
Garriock M.L., Peterson R.L. and Ackerly C.A. (1989). Early stages in colonisation
of Alliurn porutn (leek) roots by the vesicular-arbuscular mycorrhizal fungus, Glomus
versiþrme. New Phytol. lL2:85-92
Gaul H., Frimmel G., Gichner T. and Ulonska E. (1972). Efficiency of mutagenesis.
In: Induced Mutations and Plant Improvement. IAEA, Vienna. pp l2l-t39
Gianinazzi-Pearson V. and Gianinazzi S. (1983). The physiology of vesicular-
arbuscular mycorrhizal roots. Plant and Soil71: 197-209
Gianinazzi-Pearson V. and Gianinazzi S. (1939). Cellular and genetical aspects of
interactions between hosts and fungal symbionts in mycorrhizae. Genome 3l:336-341
Gianinazzi-Pearson V. and Gianinazzi S. (1992). Influence of intergeneric grafts
between host and non-host legumes on formation of vesicula¡-arbuscular mycorrhiza.
New Phytol. 120 505-508
764
Gianinazzi-Pearson V., Brazanti B. and Gianinazzi S. (1989).In vitro enhancement of
spore gerrrination and early hyphal growth of a vesicular-a¡buscular mycorrhizal
fungus by host-root exudates and plant flavanoids. Symbiosis 7:234-255
Gianinazzi-Pearson V., Smith S.E., Gianinazzî,5. and Smith F.A. (1991). Enzymatic
studies on the metabolism of vesicular-arbuscular mycorrhizas: V. Is H+ - ATPase a
component of ATP-hydrolysing enzyme activities in plant-fungus membranes. New
Phyrol. 117:61-74
Gianinazzi-Pearson V., Tahiri-Alaoui 4., Antoniw J.F., Gianinazzi S. and Dumas E.
(1990) V/eak expression of the pathogenesis related PR-b1 gene and localisation of
related protein during symbiotic endomycorrhizal interactions in tobacco roots.
Endocyto. Cell Res. 8: 177-185
Gibson S. and Somerville C. (1993). Isolating plant genes. Trends in Biotech. ll:306-313
Gilmore A.E. (1971). The influence of mycorrhizae on the growth of peach seedlings.
J. Am. Soc. Hort. Sci.96: 35-38
Giltrap N.J. and Lewis D.H. (1982). Catabolic repression of the synthesis of pectin
degrading enzymes of Suillis luteus (L. ex Fr) S.F. Gray and Hebeloma oculatum
Bruchet. New Phytol. 90: 485490
Giovannetti M. and Mosse B. (1980). An evaluation of techniques for measuring
vesicular-arbuscular mycorrhizal infection in roots. New Phytol. 84: 489-500.
Giovannetti M., Avio L., Sbrana C. and Citernesi A.S. (1993). Factors affecting
appressorium development in the vesicular-arbuscular mycorrhizal fungus Glomus
mosseae (Nicol. and Gerd.) Gerd. and Trappe. New Phytol. 123: ll5'L22
Giovannetti M., Sbrana C. and Logi C. (1994a). Early processes involved in host
recognition by arbuscular mycorrhizal fungi. New Phytol.l2T:7O3-7O9
165
Giovannetti M., Sbrana C., Avio L., Citernesi 4.S., Avio L., Gollotte 4.,
Gianinazzi-Pearson V. and Gianinazzi S. (1994b). Recognition and infection process,
basis for host specificity of arbuscular mycorrhizal fungi. In: Impact of Mycorrhizas
on Sustainable Agriculture and Natural Ecosystems. Gianinazzi S. and Schuepp H.
(eds.), Birkhauser Verlag, BaseUSwitzerland. pp 6l-72
Gish W. and States D. (1993). Identification of protein coding regions by database
similarity search. Nat. Genet . 3: 266-272
Glenn M.G., Chew F.S. and Williams P.H. (1985). Hyphal penetration of Brassica
(Cruciferae) roots by a vesicular arbuscular-mycorrhizal fungus. New Phytol. 99:
463472
Glenn M.G., Chew F.S. and Williams P.H. (1988). Influence of glucosinolate
content of Brassica (Cruciferae) roots on growth of vesicular-arbuscular mycorrhizal
tungi. New Phytol. 110: 217-225
Gollotte 4., Gianinazzi-Pearson V., Giovannetti M., Sbrana C., Alvio.L and
Gianinazzi S. (1993). Cellular localisation and cytochemical probing of resistance
reactions to arbuscular mycorrhizal fungi in a 'locus a' myc- mutant of Pisum sativum
L. Planta 191: 112-122
Gottschalk V/. and rffolff G. (1983). Induced Mutations in Plant Breeding. Springer-
Verlag Berlin Heidelberg, pp.L28-14
Govers F., Nap J-P., Moermil M., Franssen H.K., van Kammen A. and Bisseling
T. (19S7). cDNA cloning and developmental expression of pea nodulin genes. Plant
Mol. Biol. 8:425-435
Graham J.H. (1982). Effects of citrus root exudates on germination of
chlamydospores of the vesicula¡-a¡buscular mycorrhizal fungus Glomus epigaeus.
Mycologia 74:83I-835
166
Gurr S.J., McPherson M.J., Scollan C., Atkinson H.J. and Bowles D.J. (1991)
Gene expression in nematode-infected plant roots. Mol. Gen. Genet. 226l.361-366
Guttenberger M. and Hampp R. (1992). Ectomycorrhizins - Symbiosis-specific or
artifactual polypeptides from ectomycorrhizas? Planta 188: 129-136
Hall I.R. (1978). Effect of vesicular-arbuscular mycorrhizas on two varieties of maize
and sweetcorn. N.Z. J. Agric. Res. 21: 517-519
Hardie K. and Leyton L. (1981). the influence of vesicular-arbuscular mycorrhiza on
growth and water relations of red clover. I. In phosphate deficient soil. New Phytol.
89:599-608
Harley J. L. and Smith S.E. (1983). Myconhizal Symbiosis. London, New York:
Academic Press.
Harper J.F., Manney L.and Sussman M.R. (1994). The plasma membrane H+ -
ATPase gene family inArabidopsis : genomic sequence of AHA10 which is primarily
expressed in developing seeds. Mol. Gen. Genet. 244:572-587
Harper J.F., Manney L., DeWitt N.D., Yoo M.H. and Sussman M.R. (1990). The
Arabidopsis thaliana plasma membrane H+-ATPase multigene family. J. Biol. Chem.
265 :13601-13608
Harper J.F., Surowy T.K. and Sussman M.R. (1989). Molecular cloning and
sequence of cDNA encoding the plasma membrane proton pump (H+-ATPase) of
Arabidopsis thaliana. Proc. Nat. Acad. Sci. (USA) 86: I234-L238
Harrison M.J. and Dixon R.A. (1994). Spacial patterns of expression of
flavanoid./isoflavanoid pathway genes during interactions between roots of M¿dicago
trunculata and the mycorrhizal fungus Glomus versiforme. Plant J. 6:9-2O
Hayman D.S. (1983). The physiology of vesicular-arbuscular endomycorrhizal
symbiosis. Can. J. of Bot. 6l: 944-963
r67
Heckman J.R. and Angle J.S. (1987). Variation between soybean cultivars in
vesicular-a¡buscular mycorrh2al fungi colonisation. Agron. J . 7 9 : 42843O
Hepper C.M. (19S4a). Inorganic sulfur nutrition of the vesicular-arbuscular
mycorrhizal fungus Glomus caledonius. Soil Biol. Biochem. 16:669-671
Hepper C.M. (1984b). Isolation and culture of VA myconhizal (VAM) fungi. In: VA
mycorrhiza. Powell C.L. and Bagyaraj D.J. (eds.), CRC, Boca Raton. pp 95-112
Hepper C.M. (1979). Germination and growth of. Glomus caledonius spores: the
effects of inhibitors and nutrients. Soil Biol. Biochem. 11: 269-277
Heslot H. (1977). Review of main mutagenic compounds. In: Manual on Mutation
Breeding,IAEA, Vienna. pp. 51-59
Hinell M.C., Mehravaran H. and Gerdemann J.V/. (1978). Vesicular-arbuscular
mycorrhiza in Chenopodiaceae and Cruciferae, do they occur? Can.J. Bot. 56: 2813-
28t7
Holley J.D. and Peterson R.L. (1979). Development of a vesicular-a¡buscula¡
mycorrhiza in bean roots. Can. J. Bot. 57: 1960-1978
Houlné G. and Boutry M. (1994). Identification of at Arabidopsis thaliana Eene
encoding a plasma membrane H+ - ATPase whose expression is restricted to anther
tissue. Plant J. 5:3ll-317
Huynh T.V., Young R.A. and Davis R.rtV. (1984). Constructing and screening cDNA
libraries in lgt10 and tgtl1. In: DNA Cloning: A Practical Approach. Glover D. (ed.),
IRL, Oxford.pp. 49-78
Innes R.'W., Kuempel P.L., Plazinski J., Canter-Cremers H. and Rolfe B.G.
(1935). Plant factors induce expression of nodulation and host-range genes in
Rhizobiummeliloti. Mol. and Gen. Genetics 201: 426432
168
Islam A.K.M.R., Shepherd K.rW. and Sparrow D.H.B. (1981). Isolation and
characterisation of euplasmic wheat-barley chromosome addition lines. Heredity 46:
t6t-r74
Jabaji-Hare S.H., Thérien J. and Charest P.M. (1990). High resolution cytochemical
study of the vesicular-arbuscular mycorrhizal association, Glomus clarum x Allium
porrum. New Phytol. 114: 481496
Jacobs M. (1964). Test de I'activité mutagénique de la S-Bromodeoxyuridine et de la
S-iodo-deoxyuridine chez une plante supérieure Arabidopsis thaliana L., cultivée in
vitro. Lejeuni a 235: l-7
James Jr. D.W. and Dooner H.K. (1990). Isolation of EMS-induced mutants in
Arabidops¡'s altered in seed fatty acid composition. Theor. App. Genet. SO:241-245
Jeanmaire C., Dexheimer J., Marx C;, Gianinazzi S. and Gianinazzi-Pearson V.
(1985). Effect of vesicula¡-arbuscular mycorrhizal infection on the distribution of
neutral phosphatase activities in root cortical cells. J. Plant Physiol. 119: 285-293
Jende-Strid B. (1991). Gene-errzyme relations in the pathway of flavanoid
biosynthesis in barley. Theor. App. Genet. Sl: 668-674
Jende-Strid B. and Moller B.L. (1981). Analysis of proanthocyanidins in wild-type
and mutant barley (Hordeumvulgare L.). Carlsberg Res. Comm. 46:53-64
Jensen H.A. and Nittler L.W. (1971). Varietal differences among barley seedlings
grown with different nutrient solutions. Agron. J.63:714-7L7
John M.E. and Crow L.l. (1992). Gene expresion in cotton (Gossypium hirsutam )
fibre: cloning of the mRNAs. Proc. Nat. Acad. Sci. (USA) 89:5769-5773
Katz D.8., Sussman M.R., Mierzwa R.J., Evert R.}V. (1988). Cytochemical
localisation of ATPase activity in oat roots localises a plasma mebrane-associated
soluble phosphatase, not the proton pump. Plant Physiol. 86: 841-847
169
Kerr A. (1937). The impact of molecular genetics on plant pathology. Ann. Rev.
Phytopath. 25:87-ll0
Khan A.H. (1974). The occurence of mycorhhizas in halophytes and xerophytes and
of Endogon¿ spores in adjacent soils. J. Gen. Microbiol. 8l: 7-14
Kinden D.A. and Brown M.R. (1975). Electron microscopy of vesicular-arbuscula¡
mycorrhizae of yellow poplar. II: Intracellular hyphae and vesicles. Can. J. Microbiol.
2L:1768-178O
Kleinhofs 4., Sander C., Nilan R.A. and Konzac C.F. (1974). Azide mutagenicity-
mechanism and nature of mutants involved. In: Polyploidy and Induced Mutations in
Plant Breeding. IAEA, Vienna, pp. 195-199
Kleinschmidt G.D. and Gerdemann J.W. (1972). Stunting of citrus seedlings in
fumigated nursery soils related to the absence of endomycorrhizae. Phytopath. 62:
1447-1453
Koide R.T. (1991). Nutrient supply, nutrient demand and plant response to
mycorrhizal infection. New Phytol. 117: 365-386
Koide R.T. and Li M. (1990). On host regulation of the vesicular-arbuscular
mycorrhizal symbiosis. New Phytol. 114: 59-74
Koide R.T. and Schreiner R.P. (1992). Regulation of the vesicular-arbuscular
symbiosis. Ann. Rev. Plant Physiol. Mol. Biol. 43: 557-81
Koncz C., Martini N., Mayerhofer R., Koncz-Kalman 2., Korber H., Redei G.P.
and Schell J. (1939). High-frequency T-DNA-mediated gene tagging in plants. Proc.
Natl. Acad. Sci. (USA) 86:8467-8471
Koncz C., Mayerhofer R., Koncz-KalmanZ., Nawrath C. Reiss 8., Redei G.P. and
Schell J. (1990). Isolation of a gene encoding a novel chloroplast protein by T-DNA
tagging in Arabidopsis thaliana. EMBO J. 9: 1337 -1346
170
Koornneef M. (1991). Isolation of higher plant developmental mutants. Symposia of
the Society for Experimental Biology 45. Molecular Biology of Plant Development.
CBL, University of Cambridge. pp l-19
Krishna K.R., Shetty K.G., Dart P.J. and Andrews D.J. (1985)' Genotype
dependant variation in mycorrhizal colonisation and response to inoculation to of pearl
millet. Plant and soil86: lt3-125
Kucey R.M.N. and Paul E.A. (1982). Carbon flow, photosynthesis and Né fixation
in mycorrhizal and nodulated faba beans (Viciafaba L.). Soil Biol. Biochem.14:4O7-
412
Lamb C.J. (1994). Ptant disease resistance genes in signal perception and
transduction. Cell 7 6: 419422
Lamb C.J., Lawton M.4., Dron M. and Dixon R.A. (1989). Signals and transduction
mechanisms for activation of plant defences against microbial attack. Cell56: 215-224
Lambert D.H., Baker D.F. aand Cole H. (1979). The role of mycorrhizae in the
interactions of phosphorus with zinc, copper and other elements. Soil Sci. J. 43:976-
980
Lamont B. (1982). Mechanisms for enhancing nutrient uptake in plants with particular
reference to mediterranean South Africa and'Western Aust¡alia. Bot. Rev. 48: 597-689
Lamport D.T.A. (1977). Structure, biosynthesis and significance of cell wall
glycoproteins. Rec. Adv. Phytochem 11:79-115
Lander 8.S., Green P., Abrahamson J., Baarlow A and Daly M.J. (1987).
MAPMAKER: an interactive package for constructing primary genetic linkage maps of
experimental and natural populations. Genome 1: 174-181
171
Legocki R.P. and Verma D.P.S. (1980). Identification of "nodule-specific" host
proteins (nodulins) in soybean involved in the development of Rhizobium - legume
symbiosis. Cell 20: 153-163
Leong S.A and Holden D.W. (1939). Molecular genetic approaches to the study of
fungal pathogenisis. Ann. Rev. Phytopath.2T: 46348t
Levy Y. and Krikun J. (1980). Effect of vesicula¡-arbuscula¡ mycorrhiza on Citrus
jambhiri water relations. New Phytol. 85: 25-31
Lewis D.H. (1973). Concepts in fungal nutrition and the origen of biotrophy. Biol.
Rev.48: 261-278
Lewis D.H. (1975). Comparative aspects of the carbon nutrition of mycorrhizas. In:
Endomycorrhizas; F.E. Sanders, B. Mosse and P.B. Tinker (eds.), Academic Press,
London and New York. pp 119-148
Lindeberg G. and Lindeberg M. (1977). Pectinolytic ability of some mycorrhizal and
saprophytic hymenomycetes. Arch. Mycrobiol. I 15 : 9 -12
Linthorst H.J.B. (1991). Pathogenesis-related proteins of plants. Crit. Rev. Plant Sci.
10: 123-150
Litt M. and Luty J.A. (1939). A hypervariable microsatellite revealed by in vitro
amplification of a dinucleotide repeat within the ca¡diac muscle actin gene. Am. J.
Hum. Genet. 44:388-396
Lobler M. and Hirsch A.M. (1993). A gene that encodes a proline-rich nodulin with
limited homology to PSENOD12 is expressed the invasion zone of Rhizobium melílnti
- induced alfalfa root nodules. Plant Physiol. 103: 2l-30
Long S.R. (1989a). Rhizobium genetics. Ann. Rev. of Gen.23:483-506
172
Long S.R. (1989b). Rhizobium-Legume Nodulation: Life together in the
underground. Cell: 203'214
Luo D., Coen E.S., Doyle S. and Carpenter R. (1991). Pigmentation mutants
produced by transposon mutagenesis in Antirrhinummaius. Plant J. 1: 59-69
Maniatis T., Fritsch E.F. and Sambrook J. (1982). In: Molecular Cloning. Cold
Spring Harbour Laboratory, Cold Spring Harbour, New York.
Manners J.M., Davidson A.D., Jutidamrongphan'W., Mackinnon G.M. and Scott
K.J. (1938). mRNA induction in cereals by fungal pathogens. J. Cell. Biochem.
Suppl. l2c:284
Mansfield J.rW. (1983). Antimicrobial compounds. In: Biochemical Plant Pathology;
J.A. Callow (ed.), John Wiley and Sons, New York.pp237-265
Manske G.G.B. (1989). Genetical analysis of the effrciency of VA mycorrhiza with
spring wheat. Ag. Ecosys. Envir. 29:273-280
Mantis N.J. and Winans S.C. (1993). The chromosome response regulatory gene chv
I of Agrobacteriurn tumeþciens complements an Escherichia coli pho B mutant and is
required for virulence. J. Bacteriol. 175: 6626-6636
Marks M.D. and Feldmann K.A. (1939). Trichome development in Arabidopsis
thaliana. &. T-DNA tagging of the GLA,BROUSI gene. Plant Cell 1: 1043-1050
Marx C., Dexheimer J., Gianinazzi-Pea¡son V. and Gianinazzi S. (1982). Enzymatic
studies on the metabolism of vesicular-arbuscular mycorrhizas: iv.
Ultracytoenzymological evidence (ATPase) for active transfer processes in the host-
arbuscle interface. New Phytol.90: 37 43
' McGee P (1985). Lack of spread of endomycorrhizas of Centaurium (Gentianaceae).
New Phytol. 115: 451458
r73
McGonigle T.P., Miller M.H., Evans D.G., Fairchild G.L. and Swan J.A. (1990). A
new method which gives an objective measure of colonisation of roots by vesicula¡-
arbuscular myconhizal fungi. New Phytol. 115: 495-501
Menge J.4., Steirl€ D., Bagyaraj D.J., Johnson E.L. and Leonard R.T. (1978).
Phosphorus concentrations in plants responsible for inhibition of mycorrhizal
infection. New Phytot. 80: 575-581
Mercy M.4., Shivashankar G. and Bagyaraj D.J. (1990). Myconhizal colonisation in
cowpea is host dependant and heritable. Plant and Soil l2I:292-294
Miao G-H and Verma D.P.S. (1993). Soybean Nodulin-26 gene encoding a channel
protein is expressed only in the infected cells of nodules and is regulated differently in
roots of homologous and heterologous plants. Plant Cell 5:781-794
Monke G. and Sonnewald U. (1995). Elevated mRNA levels of the ribosomal protein
Ll9 and a calmodulinJike protein in assimilate-accumulating transgenic tobacco
plants. Plant Physiol. 107: l45t-1452
Morandi D., Bailey J.A. and Gianinazzi-Pearson V. (1984). Isoflavanoid
accumulation on soybean roots infected with vesicular-arbuscular mycorrhizal fungi.
Physiol. Plant Path. 24:357-364
Morandi D., Brazanti B. and Gianinazzi-Pearson V. (1992). Effect of some plant
flavanoids on in vitro behaviour of an arbuscular myconhizal fungus. Agronomie 12:
811-816
Moreley C.D. and Mosse B. (1976). Abnormal vesicular-arbuscular mycorrhizal
infections in white clover induced by lupin. Trans. B. Mycol. Soc. 67: 510-513
Morgante M. and Olivieri A.M. (1993). PCR-amplified microsatellites as ma¡kers in
plant genetics. Plant J.3: L75-182
r74
Morton J.B. and Benny G.L. (1992). Revised classification of a¡buscular mycorrhizal
fungi (Zygomycetes): A new order, Glomales, two new suborders, Glominae and
Gigasporinae, and two new families, Acaulosporaceae and Gigasporaceae. MycotÐ(on
37:471491
Mosse B. (1959). The regular germination of resting spores and some observations on
the growth requirements of an Endogone sp. causing vesicular-a¡buscular mycorrhiza.
Trans. Brit. Mycol. Soc. 42:213-286
Mosse B. (1973). Plant growth responses to vesicular-arbuscular mycorrhiza fV. In
soil given additional phosphate. New Phytol.72:127-L36
Mosse B. (1988). Some studies related to "independent" growth of vesicular-
arbuscula¡ endophytes. Can. J. Bot. 66:2533-2540
Muller K.J., Romano N., Gerstner O., Garcio-Ma¡oto F., Pozzi C., Salamini F. and
Rohde W. (1995). The barley Hooded mutation caused by a duplication in a
homeobox gene inEon. Nature 374:721-73O
Mullis K.B. and Faloona F.A. (1987). Specific synthesis of DNA in vitro via
polymerase-catalysed chain reaction. Meth. in Enzym. 155: 335-350
Murai N., Li Z.,Kawagoe Y. and Hayashimoto A. (1991). Transposition of muze
activator element in transgenic rice plants. Nuc. Acids Res. 19: 617'622
Nair M.G., Safir G.R. and Siqueira J.O. (1991). Isolation and identification of
vesicular-a¡buscula¡ mycorrhiza-stimulatory compounds from clover (Trifolium repens
) roots. App. Envir. Micro. 57:434-439
Nambura E., Naito S. and McCourt P. (1992). A mutant of Arabidopsis which is
defective in seed development and storage protein accumulation is a new abi 3 allele.
Plant J. 6:435441
Nap J-P. and Bisseling T. (1990). The roots of nodulins. Physiol. Plant. 79: 4O7'414
t75
Nemec S. and Meredith F.I. (1981). Amino-acid content of leaves in mycorrhizal and
non-mycorrhizal citrus rootstocks. Ann. Bot. 47:351-358
Nester E.W., Gordon M.P., Amasino R.M. and Yanofsky M.F. (1984). Crown Gall:
A molecular and physiological analysis. Ann. Rev. Plant Physiol. 35: 387413
Neumann P.M., Azaizeh H. and Leon D. (1994). Hardening of root cell walls: a
growth response to satinity stress. Plant Cell Envir. l7: 303-309
Nilan R.A. (1987). Trends in barley mutagenesis. In Barley Genetics V; Proceedings
of the Fifth International Barley Genetics Symposium. Okayama, Japan, Eds' Yasuda
S. and Konishi T (eds). Springer-Verlag, Berlin. pp 241-249
Nilan R.4., Kleinhofs A. and Fonzak C.F. (1931). The role of induced mutation in
supplementing natural genetic variability. Ann. New York Acad. Sci. 287: 367-384
Nye P. and Tinker P.B. (1977). Solute Movement in the Root System,
Oxford:Blackwell Scientific, U.K..
Olivier A.J., Smith S.E., Nicholas D.J.D., Wallace \V. and Smith F.A. (1983).
Activity of nitrate reductase nTrifolium subteruaneum.'effects of mycorrhizal infection
and phosphate nutrition. New Phytol. 94:63-79
Olsen O., Jakobsen K.S. and SchmelzerE. (1990). Development of barley aleurone
cells: temporal and spacial patterns of accumulation and cell-specific mRNA's. Planta
l8l:462466
Pacovski R.S. (1939). Carbohydrate, protein and amino acid status of Glycine -
Glomus - Bradyrhizobium symbioses. Physiol. Plant. 75:346-354
Pardo J.M. and Serrano R. (1989). Structure of a plasma membrane H+-ATPase gene
from the pLarúArabidopsis thnliana. J. Biol. Chem. 2e (JÐ: 8557-8562
r76
Passioura J.B. and Fry S.C. (1992). Turgor and cell expansion: beyond the Lockhart
equation. Aust. J. Plant Physiol. 19:565-576
Paula M.A. and Sequeira J.O. (1990). Stimulation of hyphal growth of the VA
mycorrhizal fungus Gigaspora margaritaby suspension culnued Puerariø phaseoloides
cells and cell products. New Phytol. lLS:69-75
Pear J.R., Ridge N., Rasmussen R., Rose R.E. and Houck C.M. (1989). Isolation
and characterisation of a fruit-specific cDNA and the corresponding genomic clone
from tomato. Plant Mol. Biol. 13: 639-651
Peters N.K. and Verma D.P.S. (1990). Phenolic compounds as regulators of gene-
expression in plant-microbe interactions. Mol. Plant-Microbe Int. 3: 4-8
Peterson R.L., Ashford A.E. and Allaway W.G. (1985). Vesicular-arbuscular
mycorrhizal associations of vascular plants on Heron Island, a great barrier reef coral
cay. Aust. J. Bot. 33:669-676
Phillips D.A. and Tsai S.M. (1992). Flavanoids as plant signals to rhizosphere
microbes. Mycorrhiza 1: 55-58
Phillips J.M. and Hayman D.S. (1970). Improved procedures for clearing roots and
staining parasitic and VA mycorrhizal fungi for rapid assessment of infection. Trans.
Brit. Mycol. Soc. 55:543-552.
Powell C.L. (1976). Development of mycorrhizal infections from Endogone spores
and infected root segments. Trans. Brit. Mycol. Soc. 66:439'445
Rafatski J.A. and Tingey S.V. (1993). Genetic diagnostics in plant breeding: RAPDS,
microsatellites and machines. Trends in Genet. 9:275'279
Rajapakse S. and Miller J.C. (1987). Intraspecific va¡iability for V.A. myconhizal
symbiosis in cowpea (Vigna unguiculata ). In: Genetic Aspects of Plant Mineral
t77
Nutrition; W.H. Gabelman and B.C. Loughman (Eds.), Martinus Nijhoff, Dordrecht,
Netherlands. pp 165-173
Reddy A.S.N., Jena P.K., Mukherjee S.K. and Poovaiah B.rW. (1990). Molecular
cloning of cDNAs for auxin-induced mRNAs and develomental expression of the
auxin-inducible genes. Plant Mol. Biol. 14:643-653
Redmond J.W., Batley M., Djordjevic M.4., Innes R.W., Kuempel P.L. and Rolfe
B.G. (1986). Flavones induce expression of the nodulation genes in Rhizobium.
Nature 323:632-635
Riggleman R.C., Fristensky B. and Hadwiger L.A. (1985). The disease resistance
response is associated with increased levels of specific mRNA's. Plant Mol. Biol. 4:
8l-86
Rolfe B.G. (1988). Flavones and isoflavones as inducing substances of nodulation.
Biofactors l: 3-10
Rolfe B.G. and Gresshoff P. (1988). Genetic analysis of legume nodule initiation.
Ann. Rev. Physiol. Plant Mol. Biol. 39: 297-319
Rossen L., Davis E.O. and Johnston A.W.B. (1987). Plant-induced expression of
Rhizobium genes involved in host specificity and early stages of nodulation. Trends
Biochem. Sci. 12: 430-433
Rossen L., Shearmen C.4., Johnston A.IW.B. and Downie J.A. (1985). The nodD
gene of Rhizobium leguminosarum is autoregulatory and in the presence of plant
exudate ionduces the nod A, B, C genes. EMBO J.4:3369-3373
Roth L.E. and Stacy G. (1989). Cytoplastic membrane systems involved in bacterium
release into soybean nodule cells as stuidied with two Bradyrhizobium japonicum
mutant strains. Eur. J. Cell Biol. 49:24-32
178
Rozycka M., Khan S., Lopez I., Greenland A.J. and Hussey P.J. (1995). A Zea
mays pollen cDNA encoding aputative actin-depolymerizing factor. Plant Physiol.
107: l01l-l0l2
Safir G.R., Boyer J.S. and Gerdemann J.W. (1971). Mycorrhizal enhancement of
water transport in soybean. Science 172: 581-583
Saif S.R. and Khan A.G. (1977). The effect of vesicular-arbuscular mycorrhizal
associations on growth of cereals Itr. Effects on barley growth. Plant and SoiI 47: 17-
26
Saiki R.K., Gelfand D.H., Stoffel S., Scharf S.J., Higuchi R., Horn G.T., Mullis
K.B. and Erlich H.A. (1988). Primer-directed enzymatic amplification of DNA with a
thermostable DNA polymerase. Science 239: 487 49I
Salganik R.I. and Dianov G.L. (1992). Molecular mechanisms of the formation of
DNA double-strandes breaks and the induction of genomic rearrangements. Mutation
Res.266: 163-170
Sanders F.E. (1975). The effect of foliar-applied phosphate on the mycorrhizal
infections of onion roots. In: Endomycorrhizas; F.E. Sanders, B. Mosse and P.B.
Tinker (eds.), Academic Press, London and New York. pp26l-276
Sanders F.E. and Tinker P.B. (1973). Phosphate flow into mycorrhizal roots. Pestic.
Sci.4: 385-395
Sanger F., Nickler S. and Coulson A.R. (1977). DNA sequencing with chain-
terminating inhibitors. Proc. Nat. Acad. Sci. (USA) 74:5463-5467
Scánnerini S. and Bonfante-Fasolo P. (1979). Ultrastructural cytochemical
demonstration of polysaccharides and proteins within the host-arbuscule interfacial
matrix in an endomycorrhiza. New Phytol.83:87-94
t79
Scannerini S., Bonfante-Fasolo P. and Fontana A. (1975). An ultrastructural model
for the host-symbiont interaction in the endotrophic mycorrhizae of Ornithogalum
umbellatum. In: Endomycorrhizas; Sanders F.E., Mosse B. and Tinker P.B. (eds.),
Academic Press, London and New York. pp313-324
Schenck N.C., Kinloch R.A. and Dickson D.W. (1975). Interaction of
endomycorrhizal fungi and root-knot nematode on soybean. In: Endomycorrhizas;
Sanders F.E., Mosse B. and Tinker P.B.(eds.), Academic Press, London and New
York. pp 605-617
Schenck N.C., Kinloch R.A. and Dixon D.rW. (1974). Interaction of
endomychorrhizal fungi and root-knot nematode of soybean. Proc. Am. Phytopath.
Soc.69: 1974
Scheres B., van Engelen F., van der Knaap van de Weil C., vab Kammen A. and
Bisseling T. (1990). Sequential induction of nodulin gene expression in the
developing pea nodule. Plant CelI2: 687-700
Sequeira L. (1984). Recognition systems in plant-pathogen interactions. Biological
Cell5l: 281-286
Serrano R. (1989). Structure and function of plasma membrane ATPase. Annu. Rev
Plant Mol. Biol. 40: 40-61
Sheahan J.J., Ribeiro-Neto L. and Sussman M.R. (1993). Cesium-insensitive
mutants Arabidopsis thaliana. Plant J. 3:647-656
Sigurbjornsson B. (1975). Improvement of barley through induced mutation. Proc.
3rd Int. Barley Genetics Symposium, Garching' pp 145-149
Siqueira J.O., Safir G.R. and Nair M.G. (1991). Stimulation of vesicular-arbuscular
mycorrhiza formation and growth of white clover by flavanoid compounds. New
Phytol. 1 l8: 87-93
180
Skipper H.D. and Smith G.W. (1979). Influence of soil pH on the soybean
endomyconhizal symbiosis. Plant and Soil53: 559-563a
Skolnick M.H. and V/allace R.B. (1988). Simultaneous analysis of multiple
polymorphic loci using amplifred sequence polymorphisms (ASPs). Genomics2:273-
279
Smith H.H. (1961). The reactor as a tool for resea¡ch in plant sciences and agriculture.
In: Programming and Utilisisation of Research reactors, Acedemic Press, London. pp
425438
Smith S.E. (1980). Myconhizas of autorophic higher plants. Biol. Rev. 55: 475-510
Smith S.E. (1995). Discoveries, discussions and directions in mycorrhizal research.
In: Mycorrhiza;Yarma A. and Hock B. (eds.), Springer-Verlag, Berlin Heidelberg
New York.pp3-24
Smith S.E. and Gianinazzi-Pea¡son V. (1988). Physiological interactions between
symbionts in vesicular-a¡buscular mycorrhizal plants. Ann. Rev. Plant Physiol. 39:
22t-24
Smith S.E. and Smith F.A. (1990). Structure and function of the interfaces in
biotrophic symbioses as they relate to nutrient transport. New Phytol. 114: 1-38
Smith S.E., Dickson S. and V/alker N.A. (1992). Distribution of VA mycorrhizal
entry points near the root apex: Is there a uninfectible zoÍre at the root tip of leek or
clover? New Phytol. I22: 469477
Smith S.E., Robson A.D. and Abbott K. (1992). The involvement of mycorrhizas in
assessment of genetically dependant efficiency of nutrient uptake and use. Plant and
Soil 146: 169-179
181
smith s.E., sr. John 8.J., Smith F.A. and Bromley J.-L. (1986). Effect of
mycorrhizal infection on plant growth nitrogen and phosphate nutrition of glasshouse-
grown Allium cepaL.. New Phytol. 103: 359-373
Smith S.E., St. John B.J., Smith F.A. and Nicholas D.J.D. (1985). Activity of
glutamine synthase and glutamate dehydrogenase in Triþlium subterraneumL. and
Allium cepaL.: effects of mycorrhizal infection and phosphate nutrition. New Phytol.
99:2ll-227
Snellgrove R.C., Splittstosser'W.E., Stribley D.P. and Tinker P.B. (1982). The
distribution of carbon and the demand of the fungal symbiont in leek plants with
vesicular-a¡buscular mycorrhizas. New Phytol. 92: 7 5 -87
Snellgrove R.C., Stribley D.P., Tinker P.B. and Lawlor D.W. (1986). the effect of
vesicula¡-arbuscular mycorrhizal infection on photosynthesis and carbon distribution
in leek plants. In: Physiological and Genetical Aspects of Mycorrhizae; Gianinazzi-
Pearson V. and Gtaninazzi S. (eds.), INRA, Paris. pp 421424
Son C.L. and Smith S.E. (1983). Myconhizal growth responses: interactions between
photon irradiance and phosphorus nutrition. New Phytol. 108: 305-314
Southern E. (1975). Detection of specific sequences among DNA fragments separated
by gel electrophoresis. J. Mol. Biol. 98: 503-517
Spanu P. and Bonfante-Fasolo P. (1988). Cell-wall-bound peroxidase activity in roots
of mycorrhizalAlliumporrum. New Phytol. 109: II9-124
Spanu P., Boller T., Ludwig 4., rWiemken 4., Faccio A, and Bonfante-Fasolo P.
(1989). Chitinase in roots of mycorrhizal Allium porrum: regulation and localisation.
Planta 177:447-455
Sprent J.I. (1989). Which steps are essential for the formation of functional legume
nodules? New Phytol. 111: L29-153
r82
Sreenivasa M.N. and Rajashekhara E. (1992). Influence of the stage of host
development and genotypes on mycorrhizal colonisation of field grown wheat. Z.
Mikrob. lM:38I-384
Stachel S.E., Messens E., Van Montagu M. and Zambryski P. (1985). Identification
of the signal molecules produced by wounded plant cells that activate T-DNA transfer
in Agrobacterium tumefaciens. Nature 3t8: 624-629
Staskawicz 8.J., Dahlbeck D. and Keen N.T. (1984). Cloned avirulence gene of
Pseudomonas syringae pv. Glycinea determines race-specific incompatibility on
Glycine max (L) Merr. Proc. Natl. Acad. Sci. (USA) 8t:6024-6028
Stegemann H. (1984). Retrospect on 25 years of cutivar identification by protein
pattern and prospects for the future. In: Biochemical tests for cultiva¡ identification;
Draper S.R. and Cooke R.J. (eds.),International Seed-testing Association. pp 20-31
Sussman M.R. and Harper J.F. (19S9). Molecular biology of the plasma membrane of
higher plants. Plant Cell 1: 953-960
Sutton J.C. (1973). Development of vesicular-arbuscular mycorrhiza in crop plants.
Can. J. Bot. 51: 2487-2493
Sutton J.C. and Barron G.L. (L972). Population dynamics of Endogone spores in
soil. Can J. Bot. 50: 1909-
Swegle M., Huang J-K., Lee G. and Muthukrishnan S. (1989). Identification of an
endochitinase cDNA from barley aulerone cells. Plant Mol. Biol. L2: 4O3412
Tagu D., Python P., Crétin C. and Martin F. (1993). Cloning symbiosis-related
cDNA's from eucalyptus mycorrhiza by PCR-assisted differential screening. New
Phytol. 125 339-343
Tester M., Smith S.E. and Smith F.A. (1987). The phenomenon of 'nonmycorrbizal'
plants. Can. J. Bot. 65: 419-431
183
Timmis J.M., Whisson D.L., Binns 4.M., Mayo M.J. and Mayo G.M.E. (1990).
Deletion mutation as ameans of isolating avinrlence genes in flær rust. Theor. Appl.
Genet. 79:411416
Tinker P.B. (1975a). Effects of vesicular-a¡buscular mycorrhias on higher plants.
Symp. Soc. Exper. Biol. 29:325- 371
Tinker P.B. (1975b). Soil chemistry of phosphorus and mycorrhizal effects on plant
growth. In: Endomycorrhizas; F.E. Sanders, B. Mosse and P.B. Tinker (eds.),
Academic Press, London. pp 353-371.
Tommerup I.C. and Abbott L.K. (1981). Prolonged survival and viability of V-A
mycorrhhizal hyphae after root death. Soil Biol. Biochem. 13:431433
Torp J. and Jorgensen J.H.. (1986). Modification of barley powdery mildew
resistance gene Ml-al2 by induced mutation. Can. J. Genet. 28:725-731
Toth R. and Miller R.M. (1984). Dynamics of arbuscule development and
degeneration in a Tnamays mycorrhiza. Am. J. Bot. 7l:449460
Toth R., Page T. and Castleberry R. (1984). Differences in mycorrhizal colonisation
of maize selections for high and low ear leaf phosphorus. Crop Sci. 24:994-996
Toth R., Toth D., Sta¡ke D. and Smith D.R. (1900). Vesicular-a¡buscular mycorrhizal
colonisation in ka mays affected by breeding for resistance to fungal pathogens. Can.
J. Bot.68: 1039-1044
Trinick M.J. (1977). Vesicular-arbuscular infection and soil phosphorus utilisation in
Lupinus spp. New Phytol. 78:297-304
Van Kammen (1984). Suggested nomeclature for plant genes involved in nodulation
and symbiosis. Plant Mol. Biol. Rep. 2: 43-45
184
Van Loon L.C., Gerritsen Y.A.M. and Ritter C.E. (1987). Identification, purification
and characterisation of pathogenesis-related proteins from virus-infected Samsun NN
tobacco leaves. Plant Mol. Biol. 9: 593-598
Van Sluys M.4., Tempé J. and Federoff N. (1987). Studies on the introduction and
mobility of the muze Activator element in Arabidopsis thaliana and Daucus carotus.
EMBO J. 6: 3881-3889
Vierheilig H. and Ocampo J.A. (1991). Receptivity of various wheat cultivars to
infection by VA mycorrhizal fungi as influenced by inoculum potential and the relation
of VAM-effectiveness to succinic hydrogena^se activity of the mycelium in roots. Plant
and Soil 133:29I-296
Wada M., Takano M. and Kasamo K. (1992). Nucleotide sequence of a
complementary DNA encoding pla^sma membrane H+-ATPase from rice (Oryza sativa
L.). Plant Physiol. 9:794-795
Walbot V. (1992). Strategies for mutagenesis and gene cloning using transposon
tagging and T-DNA insertional mutagenesis. Ann. Rev. Plant Physiol.Plant Mol.
Biol.43: 49-82
Walden R., Hayashi H. and Schell J. (1991). T-DNA as a gene tag. Plant J. L:28L-
288
V/eining S. and Langridge P. (1991). Identification and mapping of polymorphisms in
cereals based on polymerase chain reaction. Theor. Appl. Genet. 82:209-216
'Westenberg B., Boller T. and Wiemken A. (1989). Lack of arginine-storage and
polyphosphate-storage pools in a vacuole-deficient mutant (endl) of Sacchoromyces
cerevisiae. FEBS I-etters 254: 133-136
V/innacker E.L. (1987). From Genes to Clones: Introduction to gene cloning. VCH
Weinheim. ppl49-177
185
Woloshuk, C.P. and Kolattukudy, P.E. (1986). Mechanism by which contact with
plant cuticle triggers cutinase gene expression in the spores of Fusarium solani f. sp.
pisi. Proc. Nat. Acad. Sci. (USA) 83: 1704-08
Wyss P., Boller T. and Wiemken A. (1990b). Vesicular-Arbuscula¡ mycorrhizas of
wild-type soybean and non-nodulating mutants with Glomus rnossea¿ contain
symbiosis-specific polypeptides (mycorrhizins) immunologically cross-reactive with
nodulins. Planta 182: 22-26
Wyss P., Boller T. and V/iemken A. (1991). Phytoalexin response is elicited by a
pathogen (Rhizoctonia solani ) but not by a mycorrhizal fungus (Glomus mosseae ) in
soybean roots. Experientia 47 : 395401
Yanisch-Penon C., Veira J. and Messing J. (1985). Improved Ml3 phage cloning
vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors.
Gene 33: 103-119
Yoder J.I. (1990). A genetic analysis of mutations recovered from tomato following
Agrobacterium- mediated transformation with the maize transposable elements
Activator and Dis s o ciation. Theor. Appl. Genet. 7 9 : 657 -662
Young J.L., Davis E.A. and rose S.L. (1985). Endomycorrhizal fungi in breeder
wheats andTritical¿ cultivars field grown in fertile soils. Agro¡.77:219-224
Zehetner G., Frischauf A. and Lehrach H. (1987). Approaches to restriction map
determination. In: Nucleic Acid and Protein Sequence Analysis: A Practical Approach.
Bishop M.A. and Rawlings C.J. (eds.), IRL Press Ltd., Oxford'Washington D.C..
pp 147-164
Zidan L, Shavav 4., Ravina I., and Neumann P.M. (1992). Does salinity inhibit
muze leaf growth by reducing tissues concentration of essential mineral nutrients. J.
Plant Nutrit. 15: 1407-1419
186
A nYrPnd¡Y 1
Sources or Composition of Materials and Solutions
SourceEscherichia coli sfratns
HBIOlDH5a
ER1647
C600 andC600hfl
Enzvmes
: Professor A. Kerr.
: Ms. Jan Nield.
: New England Biolabs.
: Promega.
Source
Promega
Pharmacia
Boehringer Mannheim
Promega
Boehringer Mannheim
Boehringer Mannheim and Promega
homega
Boehringer Mannheim
Pharmacia
Sigma
Sigma
BRL
Promega.
Promega
Alkaline Phosphatase
E. coli DNA Polymerase I
Klenow DNA Polymerase
DNase IT4DNALigase
Restriction Endonucleases
RNase inhibitor
RNase A
RNase H
Lysozyme
Proteinase KReverse Transcriptases :
Superscript RT H-
M.MLVRTTaqPolymeruse
Radio-Nucleotides
Radio-nucleotides 1aJ2P-dCTP) were supplied by Amersham.
187
Buffers. Solutions. Microbial Growth Media and Electrophoresis Gels
The buffers etc. used for these investigations are listed hereunder in alphabetical
order along with their compositions.
Agarose gels : O.7-3glL of agarose (as indicated) in
either TAE or TBE buffer.
Alkali lysing solution : 0.5M NaOH
: l.5M NaCl
Chloroform/IAA : Chloroform/Isoamyl alcohol (24:l vlv)
Denanuing solution : 1.5M NaCl
: 0.5M NaOH
Denha¡dts Solution : 27o Polyvinylpyrolidone
: 2%oEicoll4O0
: 2%oB,ovtne Senrm Albumin
Depurination solution : 0.25N HCI
FormamideÆormaldehyde : 89pl Formaldehyde (37Vo)
: 250¡tl deionised Formamide
Hybridisation solutions :
Solution I : Prehybridisation Solution 1 plus 200p1
denatured salmon sperm (5 mg/ml),
denatured radioactive probe and 3ml of
Dextran Sulphate (25Vo w:w)
Solution 2 : Prehybridisation Solution 2 plus 200p1
of denatured salmon sperm (5 mg/ml),
denatured radioactive probe and lml of
Dextran Sulphate (25Vo w :w)
Ligase buffer : as supplied by Promega.
188
LB complete medium : 10g TryPtone
: 59 YeastExtract
: 10g NaCl
(Medium was adjusted to pH 6.8 before autoclaving.)
MOPSÆDTA Buffer (lùr) : 0.5M MOPS
: l0mlvl EDTA
(adjusted to pH 7.0 before autoclaving)
Neutralisation buffer : l.OM Tris-HCl (pH 6.5)
: 2.0M NaCl
(adjusted to pH 5.0)
Phage Buffer : 20mM Tris-HCl (pH 7.4)
: lO0rnM NaCl
: l0mlvf MgSO4
Plant Nutrient Solutions (5 or l0 times concentration as indicated)
Solution A (lOX) : 0.6M KZSO¿
: 0.2M MgCl2
: 12mM It¡InSO+
: lOmlvI ZnSO+
: 6mM CUSO4
: 1.5mM CoSO¿
Solution B (l0X) : 3mM H3BO3
: 1.5mM Na2MoO4
Solution C (5X) : 0.7M Ca(NO¡)Z
: l.4M KNO¡
Solution D (5X) : l6mMFeSO4
Solution E (5X) : 2M NH4
Phosphate Solution : lM NaH2PO¿
(All solutions were made using npH2O and stored at 4oC. Solutions A-E
were diluted to the correct concenEations when required )
189
Prehybridisation S olutions
Solution I : 3ml Denha¡dts Solution
: 6ml5x HSB
: lSml nanoPure H20
(The solution was made up fresh for each hybridisation and warmed to 65oC.
Three ml of freshly boiled salmon spefm DNA (10 mg/ml) was added just
before the prehybridisation began)
Solution 2 : 5ml Deionised formamide
: 5ml20x SSPE
: 2rnlDenhardts solution
: lml 107o SDS
: 6ml nanopure H2O
(The solution was made up and stored as 19ml aliquots at -20oC. The solution
was thawed on ice and lml of freshly boiled salmon sPerm DNA was added just
before the prehybridisation began)
Phenol : Phenol equilibrated in I x TE
and O.lVo hydroxy-quinoline
PhenoUChloroform : SOVo Phenol
: 50Vo Chloroform/Isoamyl alcohol
RNA buffer A : 294¡il,MOPSÆDTA (10x) buffer
: 706pf nanopure H2O
RNA denanuing gel : 0.5g Agarose
: 36ml nanopure H20
: 5ml MOPSÆDTAbuffer
: 9ml formaldehYde (37Vo)
(The agarose was dissolved and cooled to 60oC
before addition of the formaldehyde)
190
RNA loading buffer
SDS (107o)
SM buffer
SSC (20 x)
TAEbuffer
TBE buffer
TE buffer
V/ash buffer 1 (to 2L)
Wash buffer 2 (to2L)
: 322¡tI RNA buffer A
: 178p1 formaldehyde (37 Vo)
: 500¡tl deionised formamide
: 3mg xylene cyanol
: 3mg bromocresol green
: 250mg sucrose
: l00g SDS
: 900m1nanopure H2O
Phage buffer plus 0.17o gelatin
: 6.0M NaCl
: 0.6M Na Citrate
: 400mM Tris-HCl (pH7.5)
: lOmlvlEDTA
: 50mMNaAcetate
: 500mM Tris-HCl (pH 7.5)
: l0mlvIEDTA
: 50ornlvf Boric acid
: 10mM Tris-HCl (pH 7.5)
: ImMEDTA
: 200m1SSC (20x)
: 200m1SDS (107o)
: 1600m1nanopure H2O
: 100m1SSC (20x)
: 20ml SDS (107o)
: 1880m1nanopure H2O
: 20ml SSC (20x)
: 20ml SDS (20x)
: 1960m1nanopure H2O
V/ash buffer 3 (to 2L)
191
Miscellaneous materials.
'Geneclean'plasmid DNA purification kits were purchased fromBIO l0l Inc.
'Hybaid' oven and hybridisation bottles were provided by Dr. Peter Langridge,
Dept of Plant Sciences.
t92
An rlix 2
1
2
3
DNA Isolations from Plant Roots and Fungal Spores
This method uses a high salt buffer containing cationic detergent
cetyltrimethylammonium bromide. Use wide-bore (ie. cut-off blue) tips or glass
pipettes for transfering all DNA preparations as this is a major source of DNA
shearing. Gloves must be worn (phenol and p-mercaptoethanol are used in this
procedure).
Freeze tissue (50mg-0.259) in liquid N2 and grind to a fine powder with a
suitably sized mortar and pestle.
Add 0.5-5mt of the extraction buffer (see below) and regrind for a further l-2
minutes.
Transfer the slurry to a 15ml Corex tube (use a2rrl Eppendorf tube for volumes
less than lml) and incubate at 65"C in a water bath for 20 minutes.
Extract the proteins with an equal volume of PhenoUChloroform/IAA solution
Separate the phases by centrifugation (5 minutes/12000g) and transfer the
aqueous (upper) phase to a clean tube.
Re-extract with an equal volume of Chloroform/IAA.
Separate the phases by centrifugation (5 minutes/12000g) and transfer the
aqueous (upper) phase to a clean tube taking care not to transfer any of the
material deposited at the interphase of the two phases.
Ethanol precipitate the DNA (see 2.7.3.5) and recover by centrifugation (10
minutes/12000g).
rr)Vash the pellet 2-3 times with a 707o ethanol solution (lml) to remove
contaminating salts. Take care to remove as much of the residual ethanol as
possible before drying under vacuum for 10 minutes.
Resuspend the pellet in an appropriate volume (20-100p1) of RNAse solution
(aOpg/ml in TE).
4
5
6
7
8
9
9
193
EXTRACTION BI.JFFER
10OmM Tris-HCl (PH7.8)
50mI\4 EDTA
S0omlvf NaCl
I 7o cetyltrimethylammonium bromide
3 ¡rl of 987o þ-mercaptoethanol for each 5ml of buffer (add just prior to use)
t94
1
DNA Isolations from Plant Leaves
This protocol is based on the method used by Guidet et aI. (1991) and uses a high-salt
buffer to reduce erizyme activity and the detergent'Sarcosyl' to disrupt the cell nuclea¡
membranes during the grinding procedure. It is important to work quickly and at the
coldest temperature possible in order to minimise oxidative reactions' However, do
not add too much liquid nitrogen at the start or the 'Sarcosyl' will solidify and take 5-
l0 minutes to re-liquify. Use wide-bore (ie. cut-off blue) tips or glass pipettes for
transfering all DNA preparations as this is the major soluce of DNA shearing. Gloves
must be worn (phenol and p-mecaptoethanol a¡e used in the procedure).
Grind 1-1.5 g of plant leaf tissue in liquid nitrogen; regrind 2 ot 3 times by
adding a liffle extra liquid nirogen.
Add 45 ml of extraction buffer (see below) and ll2 weight (ie.150-750mg) of
insoluble Polyvinylpirolidine and mix for a further 1 minute while still cold.
Transfer the slurry into two 30 ml Corex tubes and centrifuge (10
minutes/10000814"c) to pellet cell debris.
Transfer the supernatant (about 20ml) to fresh Corex tubes and extract with
phenoVchloroform/IAA (see appendix 1) for at least 10 minutes. Centrifuge (5
minutes/10000gt4"C) to separate phases and transfer aqueous (upper) phases to
a new set of tubes.
Chloroform/IAA (see appendix l) extract for 5 minutes to remove residual
phenol. Centrifuge as above.
Add a l/10th volume of 3M NaAcetate to each tube then precipate the DNA by
adding an equal volume of iso-propanol.
Allow the mixture to stand at room temperature for about 5 minutes then pellet
the DNA by centrifugation (10 minutes/10000g/4'C). Wash the pellet thrice with
'70Vo ice-cold ethanol (about 10 ml). Try and remove as much of the ethanol as
possible using a fine pipette.
Resuspend the DNA in320 pl of npH2O, add 80 pl of 4MNaCl and 400 pl of
I3Vo PEG solution.
2
3
4
7
8
5
6
195
9 Leave on ice for at least 30 minutes (the longer the better) and then pellet the
DNA by centrifugation. rWash withT}Vo ethanol as above and then resuspend in
an appropriate volume of RNase solution (40pg/ml in TE).
EXTRACTION BUFFER
l00mM Tris-HCl (pH7.8)
sOmIvIEDTA
l00ml\d NaCl
47o Sarcosyl
35 ¡rl of 98Vo þ-mercaptoethanol for each 50ml of buffer (add just prior to use).
196
Ano dix 4
I
2
Plasmid DNA Isolations
Pick a single bacterial colony from plate and grow overnight in 5ml of LB buffer
containing appropriate antibiotic.
Decant 1.5m1of culture into Eppendorf tube and cenhifuge (2 minutes/4O00g)to
pellet cells.
Add l00pl of Solution 1 and thoroughly resuspend the cells by vortexing.
Incubate at room temperature for 5 minutes.
Add 200p1 of Solution 2 andmix by gentle inversion of the tube. Incubate on ice
for 5 minutes.
5. Add 150p1 of Solution 3 and vortex briefly to mix the contents of the tube
Incubate on ice for 5 minutes.
Centrifuge (5 minutes/12000g) the tube and transfer the aqueous (upper) phase
into a fresh tube containing 450F1 of phenoUchloroform/IAA (see Appendix l).
Extract for 5 minutes by forming an emulsion every minute (gently invert the
tube).
Centrifuge (5 minutesll2000g) the tube and transfer the aqueous (upper) phase
into a fresh tube containing 200p1of chloroform/IAA (see Appendix l).
Ethanol precipitate the DNA (see 2.7.3.5) and recover by centrifugation (10
minutes/12000g).
Wash the pellet 2-3 times with a 7O7o ethanol solution (0.5m1) to remove
contaminating salts. Take care to remove as much of the residual ethanol as
possible before drying under vacuum for 10 minutes.
Resuspend the pellet in an appropriate volume (50-100p1) of RNAse solution
(40pg/ml in TE).
3
4
6
7
8
9
9
9
r97
PLASMID-IS OLATION SOLUTIONS
Solution l:5OmNd Glucose
25mM Tris-HCl (pH 8.0)
IOInI\{EDTA
Solution 2: (freshly prepared for each plasmid preparation)
0.2N NaOH
l7o SDS
Solution 3:
3.0M K Acetate (hepared by adding 1l.5ml glacial acetic acid and 28.5m1 HzO
to 60ml of 5M potasium acetate)
Adjust to final pH 4.8 with galacial acetic acid.
198
1
2
Phage DNA Isolations
Core out phage plaque with a sterile Pasteur pipette and expel into 200¡tl of SM
buffer. Allow the phage to diffuse into the buffer for at least 2 hours at 4"C.
Add 100Ut of the phage solution to 400m1of competent C600 hfl cells in a large
(50m1) test tube
Allow the phage to adsorb to the cells for 20 minutes then add 5ml of LB
(+5mM CaClù and shake vigorously (200rpm) at37"C until the culture lyses
(usually 5-6 hours).
Transfer the lysate to a lOml plastic centrifuge tube and spin down the cellula¡
debris ( l0 minutes/5000g).
Remove the supernatant, add DNAse J and RNAse A to a final concentration of
lpglrnl and incubate at37"C for 30 minutes.
Add 5ml of PEG solution (see below) and allow to precipitate on ice for at least
t hour.
Pellet the phage by centrifugation (15 minutes/10000g), pour off the supernatant
and invert the tubes (on absorbant PaPer) and allow to drain for 15 minutes.
Resuspend the phage pellet in 700p1 of LB and transfer to a 1.5m1 Eppendorf
tube containing 700p1 of DEAE-52. Gently invert the tubes 20-30 times.
Centrifuge (5 minutes/12000g) to spin down the resin and transfer the
supernatant to a fresh tube. Add 13¡rl of Proteinase K (0.1m9/ml) and 32pl of
107o SDS. Mix and incubate at room temperature for at least 10 minutes.
10. Add 130p1 of 3M potasium acetate solution (see Appendix 4) and incubate at
88)C for 20 minutes then on ice for 10 minutes.
1 1. Centrifuge (5 minutes/12000g) to pellet the denatured proteins and transfer the
supernatant to a fresh Eppendorf tube. A phenoVchloroform/IAA extraction can
be incorporated at this point but it is not necessary and will reduce the final yield
of phage DNA.
3
4
5
6
7
8
9
r99
IZ. Add an equal volume of isopropanol to precipitate the DNA. the tubes can placed
on ice for t hour or snap frozen and left in liquid N2 for 5 minutes.
13. Centrifuge (10 minutes/12000g) to pellet the DNA, wash 2-3 times withTOVo
ethanol and resuspend the pellet in DNAse-free H2O or TE (30-50F1).
PHAGE PRECIPITATION (PEG) SOLUTION
20Vo poly ethyleneglYcol (w : v)
2M NaCl in SM buffer
200
A rlix 6
PCR-Primer Sequences
Lambda gt10 Primers
AT Primers
CP Primers
LA Primers
ST Primers
Rl Primer
E2 Primer
E4 Primer
5' - AGCAAGTTCAGCCTGGTTAAGT - 3'
5' - CTTATGAGTATTTCTTCCAGGGTA -3'
5'- ATGGCTTCGGTTACTTTC - 3'
5'- AGACGATGGAJU\CGAACT - 3'
5' - .A.A,GTACACCCTCAAAAGC - 3'
5'- TTGCCAGAAACAATAACC - 3'
5'- GGACGTTGTAGAACTTGT - 3'
5'- ACTGTTCCACTATTCGTT - 3'
5'- CAGCTCAACÀAGTGGTAA - 3'
5' - ACA.AAAAGGGGAATCTAC - 3'
5'- TCGTCGCTGACTTAGCTG - 3'
5' . GAATTCCAGGTAAGT - 3'
5'- GGAAITCCACCTGCA - 3'